Comments submitted to the US Census Bureau on their Proposed Residence Requirements Rulemaking for the 2030 Census.

September 29, 2026

To: Daniel Sweeney Deputy General Counsel for Economic, Statistical, and Regulatory Affairs Office of the General Counsel, Department of Commerce

Cc: George Cook Deputy Under Secretary for Economic Affairs, performing the non-exclusive functions and duties of the Director of the Census Bureau, Department of Commerce Cc: Michael Lachanski Deputy Director for Data, Policy, and Science, Census Bureau

I am writing in my personal capacity to respond to the notice of proposed rulemaking published by the Bureau of the Census, Department of Commerce, on September 10, 2026, titled Decennial Census of the Population of Americans; Proposed Residence Criteria and Proposed Regulations for Demographic Questions (91 FR 57524 and available for public comment on regulations.gov docket USBC-2026-0628.

The proposed rule modifies two parts within Title 15 of the Code of Federal Regulations. Proposed Part 60 narrows the decennial apportionment base by excluding non-citizens who lack lawful permanent resident status or whose legal status is deemed less durable or indefinite. Proposed Part 70 prohibits questions concerning race, ethnicity, or sexual orientation from appearing on decennial census enumeration questionnaires. This comment argues that neither proposals are appropriate and that the 30-day comment period is insufficient for the public to adequately respond to such precedent-breaking proposed rulemaking.

I. The Proposed Residence Criteria Conflict with the Constitutional Mandate for an Actual Enumeration

A. Constitutional Text and the Plain Meaning of “Whole Number of Persons” Article I, Section 2, Clause 3 of the United States Constitution, as amended by Section 2 of the Fourteenth Amendment, directs that Representatives shall be apportioned among the several States according to their respective numbers, counting the “whole number of persons in each State.” The framers deliberately chose the broad term “persons” rather than “citizens” or “voters” or “taxpayers.” Legislative records from the 1866 debates on the Fourteenth Amendment reveal that the Joint Committee on Reconstruction explicitly considered and rejected proposals to base congressional apportionment on voters or citizens, choosing to retain the total resident population of each state as the basis for representation.

The Constitution explicitly mandates an actual Enumeration, requiring a total count of all individuals residing within each state. In Wesberry v. Sanders, 376 U.S. 1 (1964), the Supreme Court affirmed that the constitutional debates establish that House representation must reflect the number of a State’s inhabitants. In Plyler v. Doe, 457 U.S. 202 (1982), the Court confirmed that undocumented immigrants are “persons” within the plain language and scope of the Fourteenth Amendment. In Federation for American Immigration Reform (FAIR) v. Klutznick, 486 F. Supp. 564 (D.D.C. 1980), where a three-judge district court dismissed the challenge for lack of standing while noting the unbroken historical practice of the Census Bureau in counting “all resident aliens”.

B. Historical Application of “Usual Residence” Since 1790 The proposed rule misapplies the established legal standard of “usual residence”. Ever since the First Census Act, usual residence has consistently meant the place where an individual lives and sleeps most of the time. While Franklin v. Massachusetts, 505 U.S. 788 (1992) recognized that usual residence embodies enduring ties, the Court applied that principle to the narrow question of allocating federal military and civilian personnel temporarily deployed abroad to their home states. The decision did not grant the Census Bureau authority to redefine usual residence to strip resident non-citizens living within state borders of their residency status based on immigration classification.

Across twenty-four decennial censuses over 230 years, the federal government has consistently counted all persons residing in the United States without regard to immigration status. When the previous Trump Administration attempted to exclude undocumented immigrants in 2020, multiple three-judge federal district courts held that the policy was unlawful under 13 U.S.C. § 141, The Administrative Procedures Act (APA), and the Fourteenth Amendment. Although the Supreme Court vacated Trump v. New York due to ripeness and agency uncertainty regarding implementation, the underlying statutory constraints remain intact (and in Department of Commerce v. New York the Supreme Court ruled the Secretary of Commerce violated the Administrative Procedure Act because the agency’s stated justification adding a citizenship question to the Census vis-a-vis enforcing the Voting Rights Act was pretextual). By proposing in this rulemaking in § 60.4 to count only lawful permanent residents and citizens, the Census Bureau exceeds its statutory authority under 13 U.S.C. § 141 and violates the Apportionment Clause. It’s my opinion that the proposed rule is entirely inconsistent with the Founders’ plain language to enumerate everyone who lives in the US.

C. Statistical Distortions from Proposed Data Matching The proposal to rely on administrative records to identify and exclude non-permanent residents violates basic tenets of survey methodology. Demographers and scientific advisory panels have documented that administrative files exhibit significant coverage and linkage errors, especially among mobile, low-income, and immigrant populations (see National Academies (1995) National Research Council, Modernizing the U.S. Census; GAO (2024), 2020 Census: Coverage Errors and Challenges Inform 2030 Plans | U.S. GAO). Using administrative data matching to modify enumeration figures will inject differential undercounts across jurisdictions with higher immigrant populations. Furthermore, introducing citizenship or immigration status questions onto census instruments deters overall public participation, degrading data quality for all communities (see Department of Commerce v. New York).

D. The Census and the Secretary Lack the Ability to Interpret outside of the Textual meaning of the Census Clause Following Loper Bright Enterprises v. Raimondo, courts owe no deference to the Census Bureau’s interpretation of the Census Act or the Constitution. Reviewing courts must exercise independent judgment under 5 U.S.C. § 706 to determine the meaning of governing statutes. Under 13 U.S.C. § 141, Congress mandated that the decennial census tabulate the “total population” to apportion Representatives according to the “whole number of persons in each State.” Congress has never delegated authority to the Department of Commerce or the Census Bureau to redefine “persons” or “total population” to exclude individuals based on immigration status. The term “persons” is clear and mandatory; it cannot be narrowed to “lawful permanent residents and citizens” by administrative decree.

E. Methodological Inadequacy of Relying on Tax Records to Determine Usual Residence I strongly oppose the use of administrative tax data as detailed in the proposed rulemaking in § 60.2. The proposal’s requirement that a person’s usual residence be “consistent with, and evidenced by, their tax records ( e.g., tax returns, W-2 forms)” suffers from critical methodological defects that render administrative tax data fundamentally unsuitable and not fit-for-purpose, for the 2030 residency requirements. In particular, I’m most concerned about these statistical and practical limitations:

• Construct Mismatch (Mailing Address vs. Physical Living and Sleeping Location): Tax filings collect contact and mailing addresses, not physical places of residence. Tax records reflect addresses where filers receive mail, tax documents, or refunds, which frequently differ from where individuals live and sleep most of the time. Filers commonly use PO boxes, commercial mailboxes, accountant/tax preparer addresses, business headquarters, or parental/relative residences (e.g., college students and young adults).
• Temporal Misalignment and Asynchronous Data: Tax returns represent prior-year activity and provide a single point-in-time filing snapshot, not real-time residence during the census enumeration period.
• Substantial Coverage and Filing Gaps: Administrative tax records suffer from severe non-filing and coverage biases. Low-income individuals, non-earners, informal economy workers, dependent elders and children, and rural or tribal populations without standard municipal street addressing are systematically excluded or distorted in these records.
• Linkage Error and Differential Misclassification: Using tax records to verify or impute physical presence will inject differential undercounts and misclassifications across states, violating fundamental census data quality standards and statistical best practices for administrative record linkage.

II. The Census Bureau Must Continue to Collect Race and Ethnicity Data Using the 2024 Updated Standards per Statistical Policy Directive - 15

A. Indispensability for Statutory Compliance and Civil Rights Enforcement Proposed Part 70 (§ 70.2) prohibits questions regarding race and ethnicity on decennial census enumeration questionnaires. This proposal undermines the nation’s civil rights framework. Decennial census demographic tabulations at the census block level are necessary to enforce Section 2 of the Voting Rights Act (VRA) of 1965 (52 U.S.C. § 10301). Under Thornburg v. Gingles, 478 U.S. 30 (1986), VRA claims require statistical evidence of racially polarized voting and geographically compact minority communities. Recent precedent in Allen v. Milligan, 599 U.S. 1 (2023) confirms that Section 2 redistricting cases continue to rely directly on block-level demographic data and the race data from decennial census is regularly used by federal agencies as benchmark for its programmatic activities, including the equitable allocation of trillions in annual federal spending.

Furthermore, the elimination of race and ethnicity questions in Proposed Part 70 flouts the statutory mandate of Public Law 94-171 (13 U.S.C. § 141(c)), in which Congress established a specific small-area data program to facilitate non-discriminatory legislative redistricting. Because the proposed sections conflict with the plain text of 13 U.S.C. § 141 and exceed the authority delegated by Congress (particularly in a post-Chevron world), and I believe that they must be set aside as contrary to law under 5 U.S.C. § 706(2).

B. Failure of the American Community Survey (ACS) and Administrative Records (or other surveys) as Substitutes The Census Bureau suggests in proposed section § 70.4 that race and ethnicity information can be gathered through the American Community Survey (ACS) or external administrative data. This reasoning is statistically defective from a methodological perspective. The ACS is a sample survey reaching approximately 3.5 million addresses annually; it produces sample estimates subject to margins of error and does not generate data at the census block level required for redistricting and itself requires a benchmark (hence, the decennial census). Administrative files from other agencies lack standardization, contain missing fields, and omit large segments of vulnerable populations. Eliminating race and ethnicity from the decennial enumeration eliminates the only universal, granular baseline in the federal data system - the ACS is a critical survey instrument that needs to be preserved in its entirety but it is not a census. The included Regulatory Impact Analysis admits that the removal of race and ethnicity data from the 2030 census would significantly erode the quality and utility of data:

“In addition, the quality of the estimates for the ACS and other federal surveys may deteriorate after the release of the 2030 decennial census data. Numerous key dependencies have been structured around the reliable annual release of intercensal estimates for race and ethnicity, such as population controls for major federal surveys; denominators for crucial social and economic indicators (e.g., fertility/mortality rates, cancer rates, poverty rates, unemployment rates, labor force participation, etc.); required information for state and federal grants and other funding formulas; foundational data to support planning at the local level (infrastructure, real estate, and other investments); and essential inputs for countless research projects across academia, government, and the private sector. To produce and release these data, the decennial census counts of the population by race and ethnicity are relied upon, as there are no other comparably accurate, comprehensive, and reliable sources of this information. Omitting race and ethnicity from the 2030 Census questionnaire would require the development of estimates from a base population derived from outdated and/or less accurate data and would deteriorate the accuracy of the estimation method itself.” (RIA, page 15).

C. Requisite Implementation of the 2024 OMB Statistical Policy Directive No. 15 Rather than eliminating demographic questions, the Census Bureau must implement the revised standards published on March 29, 2024, by the Office of Management and Budget. The revised SPD-15 establishes three evidence-based improvements: (1) combining race and ethnicity into a single unified question; (2) adding Middle Eastern or North African (MENA) as a distinct minimum reporting category; and (3) requiring the collection of detailed subcategories.

Empirical evaluations conducted by the Census Bureau, including the 2015 National Content Test Race and Ethnicity Analysis Report, showed that a combined question reduces respondent confusion, lowers item non-response, and decreases the number of respondents defaulting to “Some Other Race.” In addition, establishing the MENA category provides accurate data for millions of Americans whose origins were previously classified as “White” despite having distinct demographic profiles, economic, and health outcomes. Abandoning these standards violates SPD-15, and undermines the integrity of national statistical data.

III. The Truncated 30-Day Comment Period Violates the Administrative Procedure Act and Demands Immediate Extension

The Census Bureau has designated a 30-day public comment period ending on October 13, 2026. Under APA, 5 U.S.C. § 553(c), agencies must provide the public with a meaningful and adequate opportunity to participate in rulemaking. Providing a 30-day window for a regulatory proposal that overturns more than two centuries of the Constitution’s textual meaning in the Census Clause, the subsequent census counting methodology in every census, and eliminates core demographic measures is arbitrary and capricious. Executive Order 12866, Section 6(a)(1), and Executive Order 13563, Section 2(b), direct that “each agency should afford the public a meaningful opportunity to comment… with a comment period that should generally be at least 60 days.”

As federal courts recognized in FCC vs Prometheus Radio Project and Centro Legal de la Raza vs Executive Office for Immigration Review, rules that rest on data or technical information require adequate time for public inspection and critique. Demographers, statisticians, state and local officials, and civil rights groups cannot evaluate the impacts of this proposal in just 30 days - it’s a woefully inadequate window for the public to review the proposed rulemaking and respond appropriately. The Census Bureau should immediately publish a notice extending the comment period by at least 60 additional days, providing a total comment window of 90 days at minimum.

Conclusion

The Census Bureau should take the immediate actions:

• Withdraw Proposed Part 60: Proposed 15 CFR Part 60 must be withdrawn because excluding resident non-citizens from the decennial apportionment base violates Article I, Section 2, Clause 3 of the Constitution, Section 2 of the Fourteenth Amendment, and 13 U.S.C. § 141.
• Withdraw Proposed Part 70: Proposed 15 CFR Part 70 must be withdrawn because prohibiting race and ethnicity questions impairs enforcement of the Voting Rights Act of 1965 and undermines federal civil rights oversight and degrades the entire federal statistical system by eliminating a significant benchmark for race and ethnicity data.
• Adopt 2024 OMB SPD 15 Standards: The Census Bureau should implement the 2024 OMB Statistical Policy Directive No. 15 on the 2030 decennial enumeration form, using a unified race/ethnicity question that includes Middle Eastern or North African (MENA) as a minimum category.
• Grant a 60-Day Extension: The Census Bureau should immediately publish a Federal Register notice extending the public comment period by at least 60 days.

Thank you for the opportunity to respond.

Sincerely,
Christopher Steven Marcum
Former Senior Statistician and Senior Scientist, White House Office of Management and Budget
Gaithersburg, Maryland

Tags: general