What you need to know about the 2020 Census
The Trump administration’s plan to add a citizenship question on the 2020 Census was thrown into limbo Thursday by the nation’s highest court.
In its ruling, the U.S. Supreme Court said Trump’s Commerce Department has the authority to ask people on the census form if they are citizens, but it had so far failed to offer an adequate reason for doing so. The administration now has a few weeks to offer a more focused rationale in hopes of winning an 11th-hour judicial approval.
The question it wants to ask is itself simple. “Are you a citizen of the United States?” But the ramifications are huge.
The census count helps determine how federal funds from 55 major spending programs get distributed to states, communities and households annually for the next decade. In 2016, those funds totaled around $883 billion.
A skewed count also could tilt the balance on upcoming federal reapportioning of new House districts. California, the state with the highest estimated percentage of non-citizens, could lose a seat, and Texas may not get as many additional seats as expected. Montana may gain a seat.
That in turn affects Electoral College votes, which are based on each state’s count of House and Senate seats.
The census count also affects a portion of federal funding provided to states for programs such as Medicaid and food stamps. In North Carolina, the census was the basis for allocating $24 billion in federal funding to the state during 2016 — nearly $2,300 per resident that year, according to the George Washington University Institute of Public Policy.
Across North Carolina, an estimated 5 percent of residents were non-citizens between 2013 and 2017 — a figure lower than the national average, which is 7 percent, past census figures show. The census count of non-citizens includes people who immigrated to the United States without authorization as well as those granted temporary work visas.
In Mecklenburg County, there are about 100,000 non-citizens. Mecklenburg is one of three North Carolina counties — including Durham and Duplin — where nearly 10 percent of the population was not a citizen. In Wake County, it’s roughly 8 percent, or nearly 80,000 residents who were non-citizens.
In South Carolina, the census was the basis for allocating $13 billion in federal funding to the state during 2016 — nearly $2,600 per resident that year, according to the George Washington University Institute of Public Policy.
Across South Carolina, less than 3 percent of residents were non-citizens between 2013 and 2017 — a figure lower than the national average, past census figures show. The census count of non-citizens includes people who immigrated to the United States without authorization as well as those granted temporary work visas.
California and New York among others have sued to block the administration from asking the citizenship question. A lower court loss in the New York lawsuit prompted Trump’s Commerce Department to petition the Supreme Court earlier in January.
‘Are you a citizen?’
Under the Trump plan, a person living alone would be asked eight questions on the census form, including name, address, age and race. The final question will be: “Are you a citizen of the United States?”
Householders, such as parents, would be asked to answer a similar question for everyone living in their residence: “Is this person a citizen of the United States?”
Advocates for immigrants and non-citizens say the question is an attempt to intimidate those populations and keep them from filling out the census, thus improving the chances of creating more political districts that benefit Republicans.
Lizette Escobedo, census campaign director for the National Association of Latino Elected and Appointed Officials, said she fears even Latinos here legally may duck the census out of fear that immigration officials might use it against them.
“It is an attack on Latinos, on our immigrant community and on the democratic process,” Escobedo said.
Under Title 13 of the U.S. Code, the Census Bureau is required to keep respondent information confidential, and answers cannot be used against respondents in court or by government agencies such as immigration officials.
The Census Bureau will, however, be publishing data showing how many non-citizens live in each neighborhood.
Some non-citizens are unauthorized immigrants. Others are here legally as permanent residents with Green Cards, or with temporary work visas, or under other protected legal status.
Basim Elkarra of the Council on American-Islamic Relations said Muslim Americans may feel caught in a bind. If they fail to fill out the census, they are not counted, but if they fill it out and skip that question, the result could be scarier: “Someone is going to come knocking on your door, and that is going to make the process worse.”
Supreme Court challenge
The controversy stems from a decision announced by Commerce Secretary Wilbur Ross in a March 26, 2018 memo. Ross wrote that the federal government needs to know at the block level how many residents are citizens and how many are non-citizens in order to comply with the federal Voting Rights Act and “provide complete and accurate data” to protect minority population voting rights.
Ross noted that neither the Census Bureau nor stakeholders have documented that “the response rate would in fact decline materially.”
“While there is widespread belief among many parties that adding a citizenship question could reduce response rates, the Census Bureau’s analysis did not provide definitive, empirical support for that belief,” he wrote.
Nevertheless, he said, the bureau would put the citizenship question last, to reduce chances of impacting response rates.
In making the decision, Ross essentially bypassed the opinions of Census Bureau researchers. Chief scientist John Abowd, in a memo, had warned that adding the question “harms the quality of the census count.”
In another analysis, Census Bureau economist J. David Brown said the citizenship question “would lead to lower self-response rates in households potentially containing noncitizens, resulting in higher fieldwork costs and a lower-quality population count.”
The American Statistical Association and other sociologists and demographers accused Ross of abandoning scientific principles. In an amicus brief to the Supreme Court, they wrote that adding a question without careful testing “is grossly inconsistent with both statutory mandates and professional norms.”
The controversy took a recent twist when Common Cause and the New York Times published reports indicating that a now-deceased Republican operative, Tom Hofeller, promoted the citizenship question to the Trump administration as a way to tilt redistricting in Republican favor.
The administration disputes that contention, but the discovery has prompted critics to reassert the belief that the citizenship question is a political move.
States with a lot of non-citizens tended to back Hillary Clinton in the 2016 election. Of the 20 states with the highest proportion of non-citizens, 15 backed Clinton and only five —Texas, Florida, Arizona, Utah and Georgia — backed Trump.
Democrats in Congress have demanded administration officials disclose more about what went behind their decision. Trump invoked executive privilege to deny requested documents. A House committee earlier this month voted to recommend holding Attorney General William Barr and Commerce Secretary Wilbur Ross in contempt.
Republicans say Democrats are promoting conspiracies. They point out this would not be the first time that the Census Bureau has asked the citizenship question. The question was last asked of all residents in the 1950 census. Since then, the bureau has asked the question of a subset of about one-sixth the nation’s residents on what was called the decennial census “long-form,” a questionnaire that provided a deeper dive into demographic trends nationally.
The long form was discontinued after the 2000 census. But the bureau has been asking the citizenship question during an annual sample survey of three-plus million households called the American Community Survey.
State funding at risk
At the moment, demographers generally agree the citizenship question would likely suppress response, but there is no consensus on how much.
A recent Harvard study warns that a question that might seem innocuous becomes problematic “in an environment where the Trump White House has heavily primed Hispanic distrust in the government.”
It estimated the census could miss more than six million Hispanics nationally, in part because some household heads who fill out the firm will decline to list and identify everyone living in the residence, essentially hiding household members from the government.
Two other recent studies of the citizenship question on the American Community Survey suggest the question does not necessarily lead to honest or accurate answers.
A 2014 study by Penn State and Temple University demographers noted a potential problem: Some non-citizen immigrants are marking themselves as citizens in the American Community Survey. One of the study authors, Jennifer Van Hook, a former Census Advisory Board member, more recently said increasing mistrust of the government will lead to inaccurate results on the 2020 census if the citizenship question is included.
A 2018 study for the Georgetown Center on Poverty and Inequality found decreasing numbers of survey respondents have been answering the citizenship question annually between 2010 and 2016, notably among Asian Americans and Hispanics.
A McClatchy News review found that the number of foreign-born residents who didn’t answer the survey’s citizenship question rose again in 2017, the first year of the Trump presidency.
Georgetown report author Bill O’Hare told McClatchy he believes the citizenship question — along with the fact that the next census will be the first to be conducted mainly online — are two reasons the 2020 Census is the most problematic in decades.
“This is the most difficult census of my lifetime,” he said.
The upshot, according to U.C. Berkeley statistician Phillip Stark, is that the citizenship question could skew 2020 census results enough that it becomes statistically incomparable to the 2010 census.
“You are introducing a new variable making it hard to compare, and (potentially) making the census less suitable for its constitutional purpose.”
The possibility of an undercount in some states, though, is in fact good news in others. When one state shrinks in proportion to the others, it stands to lose federal funding, and vice versa.
California is at most risk. The latest American Community Survey suggests non-citizens account for 13.5 percent of that state’s population, the highest in the country, and nearly double the national 7 percent rate.
Texas, Nevada, New York, New Jersey and Florida, also have non-citizen populations notably above the national average, and face the possibility of diminished percentages of population-based federal funds.
States with lower percentages of non-citizens, such as North Carolina, South Carolina, Michigan, Pennsylvania and Ohio stand to gain funding.