New U.S. census rule could damage electoral democracy

The U.S. Census Bureau is standing on the precipice. As it heads into its 2030 decennial census, it is moving to fundamentally change how it counts the people of the United States. 

Though that may sound technical, the new rule will do historic damage to our democracy by redrawing our electoral map and weakening our ability to enforce voting rights laws. And it will undermine our country’s critical public data infrastructure. 

For these reasons, it must be abandoned. And Americans must speak out.

We have both been privileged to oversee parts of the federal statistical system. So we understand just how much our country depends on the Census Bureau doing its job independently, without fear or favor. That’s why we are so alarmed by the 2030 census rule (91 FR 57524 in the Federal Register).

This isn’t about politics; it’s about what the new rule would do to the basic integrity of the count itself and the way it’s used.

The census helps determine who represents you in Congress, how voting districts are drawn, how businesses make investment decisions and how communities plan for schools, hospitals, roads, emergency services and other public needs. 

Among many misguided changes in the proposed rule, two in particular stand out to us.

First, instead of counting all residents of our country — as it has always done — the census would now exclude noncitizens who lack a green card, including people here with lawful temporary status. Second, it would strip the race and ethnicity questions from the census questionnaire entirely, ending a practice that has existed, in some form, since 1790.

Defenders of these changes argue that they make the census “neutral” and “colorblind,” and that the count should reflect only people with sufficient “allegiance” to the country. We take the Constitution’s actual text more seriously than that. The 14th Amendment requires that the apportionment of the 435 seats in the House of Representatives be based on the “whole number of persons in each State” — not citizens and not people the administration in power deems sufficiently loyal. 

That language is simple and deliberate. It has been in place for more than 150 years because it takes the question of who counts out of partisan hands. 

Our concerns aren’t partisan talking points. They’re also shared by demographers, state and local government officials, civil rights lawyers, economists and census data users, including corporations that rely on this information every day.

Demographers point out that removing unauthorized immigrants from the apportionment count would shift real political power. In a 2020 analysis, Pew Research Center found that this unprecedented change would cost California, Florida and Texas one House seat each. That is not a rounding error. It is redrawing the congressional map based on who the administration decides gets to be counted as a person.

The NAACP Legal Defense Fund and the Leadership Conference on Civil and Human Rights have sounded the alarm for different but related reasons: Without race and ethnicity data, the tools used to enforce the Voting Rights Act and other civil rights statutes will lose their evidentiary foundation. You cannot prove a district was drawn to dilute Black or Latino voting strength if the government stops collecting the data that would reveal it.

Advocates have also made a related point: Even the debate over how to ask about race and ethnicity matters, because sloppy categories already blur meaningful distinctions. Eliminating the questions altogether doesn’t fix that problem. It just makes the undercount of vulnerable communities permanent and unmeasurable.

And the census doesn’t just affect voting. Public health researchers use its data to track disparities in maternal mortality, chronic disease and access to care. Businesses use it for workforce planning, market research and building data models for a range of investment decisions, including safe and sound AI. State and local governments use it to draw school boundaries, make economic development plans and allocate services. 

When the new rule’s critics span the Legal Defense Fund, scientific associations, state and city government associations, and even trade associations representing corporate America, that should tell you something.

During our time in office, we supported the professional judgment of our Census Bureau career colleagues. Why? Because when the public loses trust in an impartial, scientifically rigorous, objective census, self-participation rates fall and that leads to inaccuracies and higher costs. 

This proposed rule will lead to a census where people — including immigrants and mixed-status households with U.S. citizen children — are afraid to answer the door. And those who do participate will be more likely to omit residents who should be included on the form. This alone would degrade the count for everyone in those communities, citizen and noncitizen alike. 

We expect this rule, if adopted, to fail in court. But, tragically, litigation takes years, while damage to public trust happens immediately. That is why it is best to abandon the proposed rule altogether. It offers virtually no benefits, while its risks are enormous.

The public comment period on this rule ended Nov. 2, but there is still time to push back on the final rule. We urge anyone who relies on accurate, trusted census data — which is nearly everyone — to raise their concerns publicly and with their representatives in Congress. 

The Census Bureau’s credibility took decades to build. It’s supposed to be the one part of the federal government that both parties can trust to just count people, accurately, once every 10 years. That is what makes it work. 

This proposal treats that trust as expendable. We don’t think it should be that way.

---

Don Graves is the former deputy secretary of the U.S. Department of Commerce. Robert Santos is the former director of the U.S. Census Bureau.