Kriston Capps is a staff writer for CityLab covering housing, architecture, and politics. He previously worked as a senior editor for Architect magazine.
The GOP seems to be betting that damage from a major undercount will be isolated to Democratic-leaning cities. But it’s not that simple.
At least a dozen states plan to sue the Trump administration over its decision to add a citizenship question to the 2020 Census, with the attorneys general of New York and California—populous states with large immigrant populations—leading the charge. But the damage of a potential undercount won’t be confined to the coasts.
Before Commerce Secretary Wilbur Ross announced the citizenship question on Tuesday, the Democratic co-chair of the House Census Caucus had already proposed a bill to block last-minute census interference. Democrats in the Senate introduced mirror legislation to ensure that any changes to the census were properly tested before a survey. Another House Democrat from New York floated the possibility of withholding appropriations for the census.
While it’s Democrats who are erupting now, tracking populations that are hard for the census to reach reveals that the damage from an undercount could disrupt conservative-leaning states, too. Counties in Texas and Oklahoma, for example, contain some of the hardest-to-reach populations in the country, according to a mapping tool developed by the Center for Urban Research at the Graduate Center of the City University of New York in collaboration with the Leadership Conference on Civil and Human Rights. A major census undercount could jeopardize new congressional seat pick-ups anticipated by Texas, Arizona, North Carolina, and other states that have traditionally trended GOP.
And while counties with low response rates in Kentucky and West Virginia do not include large shares of immigrants—whose concerns about the 2020 Census predate the citizenship question—any census undercounts in those states would nevertheless undermine business development research and threaten billions of dollars in federal funds in those communities.
CUNY’s Census 2020 Hard to Count map shows mail-return rates to the 2010 Census as a measure of areas that required non-response followup actions during the last census. The data include measures by census tract, county, and legislative district. The map also tracks data for the share of households with access to the internet—a crucial detail for the 2020 Census, which will be the first to use the internet as the primary response mechanism. (Kentucky is just such a state, where poor access to the internet could require costly after-the-fact efforts to fill out the decennial count.)
The Census Bureau’s own Response Outreach Area Mapper also indicates that undercounting isn’t likely to exclusively punish Democrats. It reveals difficult-to-reach populations in Mississippi, Alabama, Georgia, and the Carolinas—plus battleground states like Colorado, New Mexico, and Florida. All of them could face diminished congressional counts or federal funds if their residents lose trust in the census.
In North Carolina, the census tracts in metro areas with low census participation rates (indicated below in blue) tend to be largely poor and African American. For example, the population for one tract in Forsyth County, in north Winston-Salem, is 84 percent black; more than 60 percent of the population lives below the poverty line. The census tract has low-response score—a wonky figure calculated by the bureau’s geography division—of 34 percent.
But the near suburbs to the south, which are significantly whiter and feature far lower poverty rates, also suffer from low census participation rates (in the 20s). These communities will also suffer from a census that is underfunded, hampered by a lack of leadership, and executed poorly for entirely predictable and preventable reasons. Worse still, the problems of an undercount linger for years and compound with every federal statistical survey.
Robert Shapiro, chairman of the economic and security advisory firm Sonecon and a senior policy fellow at Georgetown University’s McDonough School of Business, has outlined a set of broader funding concerns about how census-meddling could “boomerang” on red states. In a new post on the Brookings Institution’s FixGov blog, he writes:
My analysis … suggests that some 24.3 million people would have good reason to skip the 2020 Census if they believe their names and addresses could be shared with law enforcement. Moreover, because most of them are not concentrated in the big blue states, and most of the federal funding tied to the Census involves programs like Medicaid, Section 8 housing assistance, and support for school lunches, the new Ross–Sessions policy could cut federal funding to the 23 mainly red states with poverty rates above the national average.
For its part, the Trump administration has pitched the citizenship question as necessary to protect the Voting Rights Act. Civil rights advocates have scoffed at that argument, however, and some maintain instead that the administration aims to deliberately underrepresent Latino and African-American populations—two of the hardest for the census to reach. (These are also groups that tend to favor Democrats).
That sinister-intent interpretation got a boost when Steve King, Republican House representative from Iowa, retweeted a bewildering antebellum argument: Southern Democrats juiced their numbers through the “three-fifths compromise” (!), and Democrats aim to do the same today by counting undocumented residents.
Of course, the Fourteenth Amendment corrected the Constitution’s original sin with respect to citizenship, granting full status to anyone born or naturalized in the U.S., including former slaves. The Civil War resolved the matter over how much weight each person should be given for representation purposes. Today, it’s a settled question that the Constitution mandates a census that counts every person, full stop. That’s always been the case, from an originalist view.
Put it this way: The Trump administration’s insistence on a citizenship question is likely to lower response rates among minorities. (For a full accounting of exactly how, read Ari Berman’s feature story in Mother Jones.) Those vulnerable groups include immigrants and African Americans who live in large shares in blue states, and moreover, in blue metro areas within red states.
But as a partisan tool, the citizenship question is an ax, not a scalpel. The damage of an undercount will accrue to states, not just to cities. It’s not just Democrats who are turning out to oppose it: So are social science organizations and the American Statistical Association, which joined with a dozen different groups to form an initiative called Count on Stats, which opposes the citizenship question. Six former Census Bureau directors, stretching back to the 1970s, have likewise joined in the objections.
So far, Republican voices are missing from the outcry over the looming change to the census. National Review’s Jonathan S. Tobin figures that undocumented residents never fill out the form anyway—and doubles down on the unconstitutional premise that the census is meant to count only citizens. Republicans in Congress are mum about the Trump administration’s plan, even though an undercount could cost states such as Florida and Texas to lose congressional seats.
The census already has a purpose: to provide an entire count of the population, one against which all other federal counts are measured and adjusted. An untested alteration to the census could harm counts of Native Americans, Latinos, African Americans, children, younger people, and others. The consequences will fall on the nation as a whole—and in that accounting, Republicans won’t be spared.