by Amelia Kermis, MPH
Amelia Kermis, who holds a master’s degree in public health, is a former branch chief at the U.S. Census Bureau and a doctoral student at George Washington University’s Milken Institute School of Public Health.
The U.S. Census Bureau’s proposed 2030 Census rule would change how people are counted and remove race and ethnicity questions from the questionnaire. For rural communities, the risks associated with the changes are compounded by small populations, hard-to-reach addresses, and almost no local data beyond what the Census Bureau provides.
Where the new rule would miss rural residents
The proposal newly defines where you live as the place you “lawfully spent the greatest number of days” between January 3 and April 1, “evidenced by” your tax records (§ 60.2). Our last post covered the general problems with that standard. Rural America runs into most of them at once:
In 2020, the Bureau hand-delivered forms to roughly 6.8 million households, often in rural towns where mail goes only to a post office (P.O.) box. Tax returns leverage P.O. boxes, not the house where a rural resident actually resides.
Older adults living on Social Security may not file a return at all, and 17.5% of rural residents are 65 or older, compared with 13.8% in urban areas.
The 2020 residence rule counted nursing home residents at the facility and spelled out how to count people in assisted living and workers’ housing (2018 criteria, sections C.11 and C.19). The 2030 proposal drops the 2020 specifications and leaves them to the days-and-tax-records test, essentially leaving the determination to the IRS. Rural areas contain 38% of the nation’s nursing homes and 353,732 beds , and many residents file no tax return or list an adult child’s address. Workers living at rural job sites could be counted wherever their W-2 says they live.
Workers on temporary visas, including farm and food processing workers, would be left out of the apportionment count, and it's unclear how they will be included in other counts (§ 60.4).
In an urban area, a few thousand misplaced people doesn’t necessarily have a huge impact on the total. In a county of, say, 5,000, losing a few hundred residents may have a notable impact. For example, doing so can change eligibility, match rates, or competitive standing for federal programs, as the National Association of Development Organizations (NADO) has pointed out.
Why can’t rural places fall back on other data?
Census data guides funding for 69 federal programs serving rural areas, worth about $37.9 billion in fiscal year 2023. When the count is wrong, rural places have little to fall back on:
The error lasts a decade. The decennial count starts the clock on yearly population estimates, which set the weights for the American Community Survey (ACS). The Bureau’s own Preliminary Regulatory Impact Analysis (the PRIA) says that without new decennial data, “the ACS sampling weights, and therefore its population estimates, will become less reliable.” A county undercounted in 2030 carries such an error until 2040.
Rural places get older, averaged numbers. The proposed rule says communities can obtain race and ethnicity data from the ACS instead of the decennial census, but the Census Bureau’s own impact analysis says the decennial census numbers are “more reliable”. The ACS also gives rural places the blurriest picture of who they are. Counties with populations over 20,000 see new ACS numbers every year. In the 2025 ACS, four in 10 counties, those that are considered small with populations under 20,000, did not get yearly numbers and instead had to rely on five-year averages. For a rural health department tracking who is getting sick, a five-year average describes a community that may already have changed–in a manner that is not reflected in ACS numbers.
Smaller counts mean less detail. A new U.S. Department of Commerce order also changes how the Bureau protects privacy. When numbers are small enough to identify a real household, the Bureau has to “group” them into broad categories or leave them blank, and small places are hit hardest. Grouping data into larger areas “disproportionately affects rural and tribal communities,” North Carolina’s state data center warns. An undercount makes small numbers even smaller, and small numbers are the first to be withheld or aggregated to protect privacy.
Where are rural communities in the Bureau’s proposed rule?
The short answer is that they are missing and not called out explicitly. The PRIA says the Census Bureau “cannot quantify or provide a dollar value for the expected benefits or costs of the proposed rule.” While the analysis acknowledges that some areas may be disproportionately impacted, the only example it cites is the U.S. Island Areas. It also asks the public for help understanding “the various ways in which Census data is utilized across the country.” Rural commenters can help supply what the Census Bureau left out by showing how the rule would affect their communities.
What can you do?
Submit a comment by November 2. You don’t need to be a data expert, and the Hands Off Our Census comment guide walks you through it. If you live or work in a rural community, these points are worth making in your comment:
Name the data your community uses and the geography you need it for: a township, a school district, a county health department.
Ask how § 60.2 will count people with P.O. box tax addresses, people who don’t file taxes, nursing home residents, and seasonal residents.
Ask the Bureau to add rural areas without one-year ACS data to its distributional effects analysis, and to estimate what the rule would cost them.
Tell the Bureau what happens to your community when its numbers are wrong for ten years.