Proposed changes to the 2030 Census are throwing another curveball into the apportionment horse race.
The Trump administration proposed changes, announced last month, excluding some noncitizens from the count and removing questions about race and ethnicity. This rule change, in addition to violating the mandate of the Census, could hurt Republicans down the line.
As population forecasts ahead of the 2030 Census tend to favor red states, the proposed rule change could lead to an undercount, adding further complexity to the congressional apportionment process.
“The population of the United States is shifting to red states,” said Adam Kincaid, the president of the National Republican Redistricting Trust. “That’s just generally the case. Whether it’s these states or other red states that would gain, it’s still red states that are gaining.”
The proposed rule would exclude from the Census undocumented immigrants and noncitizens ”whose legal status is less durable and indefinite in length than lawful permanent resident status.” The Census would continue to count lawful permanent residents.
It’s not clear how congressional apportionment would be affected. A model from The New York Times projected Florida, Rhode Island, and Texas would receive fewer seats than previously forecasted under the rule, while Minnesota, Ohio, and Oregon would receive more. A draft map from Jonathan Cervas, a redistricting and apportionment expert at Carnegie Mellon University, forecasts the rule would negatively affect apportionment in California, Illinois, and Texas but could benefit Michigan, Tennessee, and Wisconsin. Kincaid’s map has eight states losing seats—chiefly California and Texas—and 11 states gaining a seat.
Most projections rely on data from the American Community Survey, which “is an unreliable estimate of noncitizens,” according to Cervas. Data on specific types of noncitizens that distinguishes between lawful permanent residents and undocumented immigrants is even more spotty.
Even without the proposed changes, the Census is highly sensitive. Based on data from the 2020 Census, New York lost a congressional district in 2024. It was just 89 residents short of keeping the district.
“Small changes to the actual numbers can reflect significant changes in the actual apportionment count,” Cervas said.
When it comes to congressional apportionment, everything is relative. Following the 2020 Census, California lost a congressional seat for the first time in state history. Its population had still increased, but at a lower rate than in past periods.
“All it does is rebalance things based off of citizenship, functionally.” Kincaid said. “If you think that people should be elected in relatively equally populated districts, then it would make sense to have apportionment based off of citizenship and also have redistricting based off of citizenship.”
The rule change is legally dubious. State attorneys general filed a lawsuit to block the Trump administration’s plan to add a citizenship question to the 2020 Census on the grounds that the Constitution requires the Census to count both citizens and noncitizens. The 14th Amendment states that representatives should be apportioned “counting the whole number of persons in each state.”
The Supreme Court ruled 5-4 in 2019 that the administration could not add the citizenship question, saying the government has the right to do so but needs to properly justify the change to the Census Bureau.
The rule lays out multiple options for how it would identify noncitizens to exclude from the count, such as adding a citizenship question or utilizing administrative records, including data from federal, state, and local agencies, to determine individuals’ legal status.
The next step after apportionment would be redistricting, which would need to account for the lower count of residents in major cities.
“You have to add people from further away, which has a dilutive effect on the people who live in [cities],” Cervas said.
This change, while not necessarily tilting apportionment in favor of Republicans, could yield individual districts more favorable to the GOP.
“If you’re drawing on the basis of citizen voting-age population, it gets harder for Democrats in certain areas of the country to draw as many Democratic districts as they’re drawing now,” election lawyer Jason Torchinsky said.
The Supreme Court ruled in 2016 that states may count all residents, regardless of eligibility to vote, when drawing districts, but it did not rule on whether states could count using other figures—for instance, only counting citizens and legal permanent residents.
The Equal Protection Clause prohibits intentional racial discrimination, which becomes much harder to prove without Census data on race and ethnicity.
“To not collect race and ethnicity data is to then make it completely impossible to ever meet that evidentiary standard,” said Meeta Anand, senior director of census and data equity at the Leadership Conference on Civil and Human Rights.
The Supreme Court weakened the Voting Rights Act in an April ruling but kept in place the core argument banning racial discrimination. However, alleging racial discrimination in the drawing of congressional districts becomes more difficult without census data showing demographic data.
“Most of our Voting Rights Act analysis comes from the Census,” said Chris Fowler, a professor of geography and demography at Penn State University. “It’s the only question that gives us the level of precision where we can confirm or deny the racial aspects of these kinds of problems."
Proponents of the change argue shortening the Census will increase overall response rates.
“By taking off the demographic information, I think there’s a reasonable argument to be made that a shorter questionnaire is going to lead to a higher response rate,” Torchinsky said.
