The Politics of the 2020 Census

May 23, 2019
10:31 AM
Originally published at La Voz de Esperanza

Editor’s Note: This piece was originally published by La Voz de Esperanza. The author has granted us permission to republish his work here.

How do you stop a locomotive from gaining speed and power?

The Latino population is that engine—the one that has propelled the U.S. population over the last several decades. Despite accounting for less than 4 percent of the nation’s population in 1970, Latinos would go on to account for almost two of every five of the nearly 122.7 million persons added to the U.S. population between 1970 and 2017. In Texas, while there were approximately 4.5 whites for every 1 Latino in 1970, Latinos are projected to outnumber whites by 2022.

Awesome numbers, indeed. The kinds of figures that make many whites and the largely white Republican Party very uneasy.

Republicans, especially in Texas, have used an arsenal of ploys to slow the Latino locomotive and stall the demographic destiny that looms near. Republicans have employed the specter of voting fraud nationally and locally to keep people of color and the poor away from the voting booth through the creation of confusion, fear, and threats. Voter ID laws and efforts to purge voter rolls have been commonly used. Republicans have also used gerrymandering tactics to draw ill-shaped maps to maintain their political power. Let’s not forget mass-incarceration policies that ensnared people of color, particularly African Americans, in the criminal justice system, disenfranchising many in the process.

Enter the decennial census. The primary function of the decennial census is to count the population in order to allocate U.S. House seats and hundreds of billions in federal dollars to states and communities. House seats are apportioned every ten years with fast-growing states gaining seats formerly held by states with little growth or decline. The 2010 census resulted in Texas being the big winner, acquiring four additional House seats, thanks largely to its growing Latino population. The Census Bureau recently estimated that $675 billion in federal funds were distributed across 132 programs in 2015.

Back room talks between Secretary of Commerce William Ross, who oversees the U.S. Census Bureau, and white nationalist and President Trump’s former chief strategist Steven Bannon birthed the idea of adding a citizenship question to the 2020 census. Just about two years before the 2020 census, Ross directed the Census Bureau to include the question on the 2020 census. Subsequently, Ross huddled up with the likes of Kris Kobach, then-Secretary of State of Kansas and head of a now-defunct Trump commission on voter fraud, and former Attorney General Jeff Sessions to devise a rationale for including the citizenship question in the upcoming census.

Ross tried to justify the inclusion of the citizenship item in order to assess voter fraud. Keep in mind that the last time that citizenship was asked in a decennial census was in 1950. The citizenship question has been asked only of a sample of the U.S. population from the 1960 to 2000 censuses and has been part of the annual American Community Survey since 2001. The citizenship item has not been used in the decennial census —aside of its use solely in a sample— because of evidence showing that it deters non-citizens from participating in the census. In fact, if Ross and his cronies are so concerned about assessing voter fraud, it can be done with data from the annual surveys of the American Community Survey. A lower court ruled against the inclusion of the citizenship item in the 2020 census and noted that Ross had acted “arbitrarily and capriciously” in deciding to add it to the census. With the Trump administration appealing the decision, the case is now before the Supreme Court, where it was recently heard and a decision will be made by the July 1 deadline when the Census Bureau will be printing the 2020 census.

The real reason for including the citizenship question is to scare off people who are not U.S. citizens and citizens who live in homes with non-citizens from participating in the census, resulting in a significantly higher than usual undercount of the U.S. population. In the process, people who are not counted do not figure into the allocation of U.S. House seats as well as in the distribution of federal funds.

Republicans will likely pursue even more draconian measures, as they have sought to eliminate the representation of noncitizens and children in the formation of political districts. The Evenwel v. Abbott case, which went all the way to the Supreme Court, sought to determine congressional districts on the basis of the number of persons 18 and older who are U.S. citizens rather than all persons. The Supreme Court ruled against Evenwel in April 2016. It is only a matter of time that Republicans will try again.

As such, the citizenship question is part of the Republican subterfuge to minimize the political power of Latinos and other people of color and, in the process, to maintain white Republican dominance. Never mind that the U.S. Constitution requires that everyone in the country, regardless of citizenship status, be counted. The ideal of democracy becomes collateral damage.

I use data from the 2017 American Community Survey to assess the impact of the representation of Latinos, whites, and the overall population across states if only U.S. citizens were to be represented. In this scenario, 21 states and the District of Columbia would experience population decreases of at least 5 percent. However, the biggest losers would be Texas along with California, Nevada, New Jersey, and New York with population reductions of at least 10 percent. California would lose 13 percent of its total with Texas coming in second with a 10.7 percent decline. Under the scenario of only citizens counted, Texas would lose slightly more than 3 million inhabitants, dropping from a total of 28.3 million residents to 25.3 million citizens.

If only citizens were represented, Latinos would be the big losers and whites the big winners. Thus, in the present situation where everyone is represented, Latinos account for 39.4 percent of the Texas population, with their percentage share dropping to 35.2 percent when only U.S. citizens are represented. In contrast, the percentage share of whites ascends from 41.9 percent to 46.1 percent, respectively.

Texas A&M University demographer, Dudley Poston, estimates that if all persons who are not U.S. citizens participated in the 2020 census, Texas would gain three House seats, but the state would end up with only two new House seats if only half of all non-citizens participated in the census. There is also the little detail of lucrative federal funds Texas would forego.

The Republican power broker trio —Governor Greg Abbott, Lieutenant Governor Dan Patrick, and Attorney General Ken Paxton— do not defend the people of Texas from the dire consequences of losing a House seat and federal dollars if non-citizens do not participate in the census. It is clear that for them, party politics trumps representation of its populace.

In sum, the citizenship question is another effort to slow the Latino population engine and forestall the demographic reality. Regardless of the Supreme Court ruling, it is of utmost importance that we fight the power structure by being counted in the 2020 census and voting to support politicians who are looking out for our interests rather than those who are more interested in cozying up to the guy in the White House.


Rogelio Sáenz is a professor in the Department of Demography at the University of Texas at San Antonio.