The House of Representatives. (National Archives)
The House of Representatives. (National Archives)

Since the first census of the United States in 1790, counts that include both citizens and noncitizens have been used to apportion seats in the House of Representatives, with states gaining or losing based on population change over the previous decade. If unauthorized immigrants in the U.S. were removed from the 2020 census apportionment count – which the White House seeks to do – three states could each lose a seat they otherwise would have had and three others each could gain one, according to a Pew Research Center analysis based on government records.

If unauthorized immigrants were excluded from the apportionment count, California, Florida and Texas would each end up with one less congressional seat than they would have been awarded based on population change alone. California would lose two seats instead of one, Florida would gain one instead of two, and Texas would gain two instead of three, according to analysis based on projections of Census Bureau 2019 population estimates and the Center’s estimates of the unauthorized immigrant population.

How we did this

This blog post explores the role of the nation’s unauthorized immigrant population in apportionment of congressional seats. Every decade, the U.S. Census Bureau conducts a count of all people living in the country, which is then used to distribute seats in the House of Representatives to the states. The analysis in this blog post is based on projections using Census Bureau population estimates, Pew Research Center estimates of the size of the unauthorized immigrant population and established formulas for assigning congressional seats.

The Method of Equal Proportions assigns congressional seats to states based on their populations after each state is given their first seat. The method requires 50 state population figures and assigns seats sequentially; it stops after the 435th seat is assigned. Our population figures for 2020 are based on the Census Bureau’s official population estimates for 2018 and 2019 projected to April 1, 2020. We use these for our baseline apportionment.

The Pew Research Center has published estimates of the unauthorized immigrant population for states through 2017. The estimates for 2016 and 2017 show very little change, and external indications suggest few changes since then. Accordingly, we use our 2017 estimates for 2020 and subtract them from the total to provide the populations for an apportionment which excludes unauthorized immigrants.

Projected change in congressional seats after 2020 census

Alabama, Minnesota and Ohio would each hold onto a seat that they would have lost if apportionment were based only on total population change. Alabama filed a lawsuit in 2018 seeking to block the Census Bureau from including unauthorized immigrants in its population count.

In addition to these states, 11 more would gain or lose seats based on population change alone, whether unauthorized immigrants are included or excluded. Five states would gain one seat each: Arizona, Colorado, Montana, North Carolina and Oregon. Six states would lose one seat each: Illinois, Michigan, New York, Pennsylvania, Rhode Island and West Virginia.

The apportionment of seats in Congress is required by the U.S. Constitution, which says that the census will be used to divide the House of Representatives “among the several States according to their respective numbers, counting the whole number of persons in each State,” except for enslaved people, who, until the late 1800s, were counted as three-fifths of a person, and certain American Indians. The 14th Amendment eliminated the partial count of enslaved people, and the total American Indian population was added later to congressional reapportionment calculations. The number of seats in the House was fixed at 435 following the 1910 census. Each state gets one seat, and the remainder are assigned according to a complex formula based on relative population size.

The census count includes everyone living in the United States, except for foreign tourists and business travelers in the country temporarily, according to Census Bureau rules. For apportionment purposes since 1990, military and civilian federal employees stationed abroad and their dependents are counted as living in a state if they provided a state address in their employment records. The District of Columbia, Puerto Rico and U.S. Island area populations are excluded from the apportionment total because they have no voting representation in Congress.

Federal law requires the population totals from the decennial census be delivered to the president nine months after Census Day, meaning Dec. 31, 2020. The Census Bureau has requested Congress extend the deadline to April 30, 2021, due to the coronavirus pandemic, although the White House reportedly may push for a “timely census” fueled by $1 billion in additional funding. States would redraw congressional district boundaries to fit the new totals. The results would take effect for the Congress that meets in 2023.

In his memorandum announcing a new policy “to the extent practicable” in how congressional seats are divided up, President Donald Trump asserted that the president has discretion to decide who is considered an inhabitant of the U.S. for apportionment purposes. Some of the same groups that successfully challenged the White House attempt to add a citizenship question to the census last year said they also would sue to block any change in apportionment policy. Democrats announced they would hold an emergency congressional hearing to respond.

The Census Bureau does not regularly publish counts or estimates of unauthorized immigrants, although the Department of Homeland Security has done so. Last year, after the U.S. Supreme Court ruled against including a question about citizenship on the 2020 census, the president ordered the Census Bureau to assemble a separate database, using other government records, on the citizenship status of every U.S. resident. This has also been challenged in court.

The Center’s analysis relies on assumptions about populations to be counted in the 2020 census and estimates of unauthorized immigrants. The actual figures used for apportionment will be different from these, and so the actual apportionment could differ regardless of whether unauthorized immigrants are excluded from the apportionment totals.

Jeffrey S. Passel  is a senior demographer at Pew Research Center.
D’Vera Cohn  is a former senior writer/editor focusing on immigration and demographics at Pew Research Center.