Count educators as part of the population taking a keen interest in a major U.S. Supreme Court case about whether President Donald Trump’s administration properly added a question about U.S. citizenship to the 2020 census.
“We are significantly concerned that big-city students would face horrific consequences from adding a citizenship question to the census,” said Julie Wright Halbert, the legislative counsel of the Council of the Great City Schools, the coalition of 74 of the nation’s largest urban districts. “There is a potential for a misallocation of huge sums of money.”
The council is among several education groups that have filed friend-of-the-court briefs in Department of Commerce v. State of New York (Case No. 18-966), which is set to be argued on April 23.
The arithmetic behind the issue is this: The decennial census is the foundation for allocation of billions of dollars of federal aid to states and localities, based on formulas involving population and poverty. In education, that includes:
• The National School Lunch Program, for which census figures are used to help distribute some $19 billion in annual aid;
• Title I compensatory education aid to school districts, at $15.8 billion;
• Grants to states under Part B of the Individuals with Disabilities Education Act, at $12.3 billion; and
• The Head Start preschool program, at $8.5 billion.
A federal district judge in January found that adding a citizenship question to the census would lead to a decline in the response rate of at least 5.8 percent of households with at least one noncitizen and would also likely result in an undercount of Hispanic households.
The Great City Schools council says in its brief that a 5.8 percent undercount of households with one noncitizen would—by itself—result in a “misallocation” nationwide of $151.7 million in Title I funds alone.
And while the impact of a census undercount might be more pronounced for urban districts, which have higher proportions of students in poverty, groups representing other sectors of K-12 education have also filed briefs in the case.
“Our interest is ensuring that the census count is accurate,” said Francisco M. Negrón Jr., the general counsel of the National School Boards Association, which filed its own friend-of-the-court brief along with groups representing superintendents, principals, and school business officials.
The NSBA brief says some 5.9 million children live in a household with at least one undocumented parent.
An accurate count “is especially important for public schools, who serve all students, regardless of immigration status,” said Negrón.
‘Foreigners Not Naturalized’
The nation’s intense debate over immigration policy is a not-so-subtle subtext to the census case, though the Trump administration’s stated rationale for the citizenship question centers on voting rights, not immigration enforcement.
The census first asked about “foreigners not naturalized” in 1820, and a citizenship question was included on all but one decennial enumeration from then through 1950. The citizenship query continued on the “long form” questionnaires that prevailed for several census cycles after 1950, and the Census Bureau now inquires about citizenship on its American Community Survey, which annually asks detailed demographic questions of a sample of the population.
U.S. Secretary of Commerce Wilbur L. Ross Jr. announced in March 2018 that he had decided to add the citizenship question in response to a request from the U.S. Department of Justice, which he said had urged the move to gain more “granular” data to better enforce the Voting Rights Act of 1965.
But legal challenges, brought by immigration groups and 18 states led by New York, turned up evidence that Ross had consulted the White House about the move and had decided to add the question well before the Justice Department made its request. Evidence also shows that Ross pushed ahead despite objections from Census Bureau experts. They had concluded that a citizenship question would depress response rates, increase costs, and result in lower-quality citizenship data than would be available by other means, such as the American Community Survey.
In January, a federal district judge in New York City ruled that Ross’ move to add the citizenship question over the objections of Census Bureau staff members violated the Administrative Procedure Act because the secretary’s rationale was pretextual and he acted in an arbitrary and capricious manner.
U.S. District Judge Jesse M. Furman said that among the “concrete harms” of adding the question was the likely reduction in Title I funding to several states that had challenged the secretary’s action.
The judge cited an affidavit in the case by Andrew Reamer, a research professor at the George Washington University Institute of Public Policy, who detailed the ways census data are used to allocate federal funds, including Title I education aid.
Reamer said in his declaration that Title I serves some 25 million children in more than 80 percent of the nation’s school districts. At least eight states would be at risk of losing Title I funds if the citizenship question is included, he concluded.
Printing Deadline Looms
The Supreme Court agreed to the Trump administration’s request to take up the case in its current term, bypassing review by a federal appeals court, because the Census Bureau must finalize its 2020 questionnaire for printing by the end of June.
In March, a federal district judge in San Francisco ruled in a separate suit that not only did Ross’ action violate the APA, it also violated the enumeration clause in Article I of the Constitution, which calls for an actual count of the population every 10 years.
The justices agreed to consider the constitutional question as part of the case in an apparent effort to resolve as much legal uncertainty as possible before the census must move forward.
U.S. Solicitor General Noel J. Francisco, in a brief defending the commerce secretary’s action, said “it simply cannot be arbitrary and capricious ... to reinstate to the decennial census a question whose pedigree dates back nearly 200 years.”
And “the decennial census has always included demographic questions notwithstanding their potential impact on response rates,” Francisco said.
A group of 17 states carried by Trump in the 2016 election filed a friend-of-the-court brief in support of the citizenship question. The brief argues that the predicted reduction in response rates rests on “numerous unwarranted assumptions” and that there is unlikely to be any “meaningful undercount.”
Two of those states, Florida and Texas, are on Reamer’s list of those likely to face a reduction in Title I funds.
Multiple Education Briefs
Christopher J. Hajec, the litigation director of the Immigration Reform Law Institute, a Washington group that also filed a brief in support of the citizenship question, said there is nothing wrong with the query.
“The census is supposed to count everybody who is a member of our political community,” Hajec said. “It seems eminently reasonable that a census that asks about race and all sorts of other things could ask about citizenship.”
Some 36 friend-of-the-court briefs have been filed in support of the challengers of the citizenship question, and most of those represent multiple groups or government agencies.
Five of the briefs represent education groups or were signed by one or more school districts.
A brief joined by the Los Angeles and Santa Clara, Calif., school districts argues that immigrants residing in those communities have reasonable fears that their responses on census forms might make their way to federal immigration authorities.
“Over the last two years, Santa Clara and Los Angeles residents have witnessed stepped-up immigration raids, and ICE [U.S. Immigration and Customs Enforcement] has arrested immigrants in courthouses and while dropping off their children at school,” the brief says.
A brief filed by the KIPP Foundation, which operates charter schools in many minority neighborhoods, argues that the census already significantly undercounts “immigrant communities of color.”
The Council of the Great City Schools used methodology similar to Reamer’s to estimate how much in federal dollars some of its member districts would lose based on a 5.8 percent undercount of noncitizen households. By the council’s estimation, that would mean $10 million less per year for the New York City school system, $3.6 million less for the Los Angeles schools, and $1 million less for the Dallas school system.
And because most districts receiving Title I funding operate so-called schoolwide programs, which allow the federal funds to be used to serve all students in a school, a census undercount would harm a broader range of students than just those below the poverty line.
Furthermore, the impact of the undercount would be compounded over 10 years, the council argues.
“The decennial census is the base,” said John W. Borkowski, a partner with the law firm Husch Blackwell in Chicago, who helped write the council’s brief. “If you start with the base being wrong, you’re going to have bad figures for a decade.”
