Joe Biden has been declared the winner of the 2020 presidential election and is scheduled to be sworn into office on Jan. 20, 2021. Ten days before Biden’s inauguration, President Donald Trump is scheduled to send a report to Congress that contains the number of people living in each state and indicates how many seats in the House of Representatives to which each state is entitled. On Monday, Nov. 30, the Supreme Court will hear oral argument in Trump v. New York, a challenge to the Trump administration’s plan to exclude people who are in the country illegally from the state-by-state breakdown used to allocate seats in the House.
If the court upholds the plan and the administration is able to implement it before leaving office, the new method of apportioning House seats could shift political power away from states with large immigrant populations and toward states with fewer immigrants.
The Constitution requires a census every 10 years to determine the population of the United States, which is then used to divide up seats in the House of Representatives. Under federal law, the secretary of commerce – to whom Congress has delegated the power to conduct the census – must send a report to the president by Dec. 31 containing “the tabulation of total population by States … as required for the apportionment” of House seats. Then, by Jan. 10, the president must send a report to Congress that contains “the whole number of persons in each State,” as determined by the census, along with “the number of Representatives to which each State would be entitled.” Throughout American history, the population numbers used to allocate House seats have encompassed all persons living in each state, regardless of their immigration status.
The dispute now before the court arose in July 2020, after Trump announced that the total population used to calculate the number of representatives for each state would not include people who are living in the country without authorization. In a memorandum, Trump instructed Secretary of Commerce Wilbur Ross to give him two sets of numbers for each state: the total population as determined by the 2020 census; and the total population as determined by the census, excluding – “to the extent practicable” – people who are not in the country legally. That second total would then become the “population base” used for congressional apportionment.
The Supreme Court ruled in 2019 that the Trump administration could not include a question about citizenship on the 2020 census. Two weeks later, Trump directed other federal agencies to provide the Census Bureau with records that he said would produce an accurate count of noncitizens living in United States. The Census Bureau has said that it is studying how it can use “administrative records” to comply with Trump’s July 2020 memorandum and provide two separate population counts.
Two groups of plaintiffs quickly went to court to challenge Trump’s new policy. One set of challengers was made up of state and local governments, led by New York; the other set was a group of nonprofits that work with immigrant communities. They argued that the memorandum violated the Constitution and federal law.
A special three-judge district court, which federal law requires for challenges to the constitutionality of the apportionment of congressional districts, blocked officials in the Department of Commerce and the Census Bureau from including any information about the number of people in the country illegally in their December report to the president. The district court agreed with the challengers that Trump’s July 2020 memorandum violates federal law because it would require seats in the House to be allocated based on data that did not come from the 2020 census. Moreover, the court added, the president does not have the option to exclude people who are not in the country legally from the total population count. The Trump administration appealed to the Supreme Court, which agreed on Oct. 16 to expedite its review of the case.
Preliminary issues: Standing, mootness and alleged harm
Even before the justices can reach the merits of the challengers’ argument, they must settle a threshold question: whether the case is properly before the justices at all. In the district court, the challengers argued that they have the legal right to sue – known as standing – because the memorandum harms them in two ways. First, they argued that the memorandum, if implemented, would have the potential to reduce congressional representation for certain states. Second, they said the memorandum had a “chilling effect” on participation in the census, causing immigrant communities to be undercounted.
The district court said the first theory was likely “too remote and hypothetical,” because it was too early to know whether any states would actually lose House seats as a result of the memorandum. But the district court accepted the second theory. The very existence of the memorandum, the court found, was deterring some immigrants (both citizens and noncitizens) from responding to the census, either because they feared that census data “will be used in immigration enforcement” or because “they perceive[d] their participation as ultimately futile” in light of the memorandum. The resulting undercount of immigrant populations, the district court concluded, risks harming the challengers through reduced representation and reduced federal funds that are tied to census data.
In a brief to the Supreme Court, the Trump administration contended that the case is now moot – that is, there is no longer a live controversy for the justices to decide. Any injury from a “chilling effect” on census participation can no longer support the challengers’ claim, the government said, because the collection of census data ended in mid-October. Even if there had been a chilling effect, the government observed, barring the secretary from implementing the memorandum now won’t do anything to prevent it.
The challengers responded that, as a result of the memorandum, there was “widespread confusion” among immigrants about whether they should participate in the census, reducing response rates and prompting the nonprofit groups that are challenging the memorandum “to divert substantial resources” to encourage participation. Even though the data collection effort has finished, the challengers continued, the case is not moot because of the possibility that a similar dispute could arise in the future. The census data collection takes eight or nine months, the challengers pointed out, which is not enough time to resolve a lawsuit claiming that a policy chills participation in the census. In similar circumstances, the Supreme Court has allowed seemingly moot claims to proceed if they are “capable of repetition” and their short-term nature would make them effectively unreviewable.
The challengers also argued that the potential harm from a loss of congressional representation is not too speculative to establish standing. The whole point of the memorandum is to take away seats in the House from states that have, relatively speaking, more residents who are not in the country legally. This is not a theoretical harm, the challengers told the justices, because the Trump administration has confirmed that it intends to implement the policy outlined in the memorandum. If it does, there is “more than a substantial risk” that at least two states – California and Texas – will lose seats.
The state and local governments contended that they have a right to sue for the additional reason that removing people who are not in the country legally from the population count could affect federal funding based on the census. They observed that the Trump administration has “conspicuously not committed to insulate” the state and local government challengers “from any funding injury.”
Substantive issues: Executive discretion and the meaning of “persons in each state”
The Trump administration argued that the district court’s ruling is also wrong on the merits. Congress, it contended, gave the secretary of commerce discretion to determine how to conduct the census, including how to count the number of people in each state. Using that discretion, the administration continued, the president can decide to instruct the secretary to do exactly what he has done here – provide two different population counts – to be used for his report to Congress. No one – either the district court or the challengers – has identified any laws or cases that suggest otherwise, the administration suggested.
The administration maintained that there “is ample historical and structural evidence” indicating that the population count does not need to include people who are living in the United States without legal authorization. The reference in the Constitution and federal law, the administration explained, to the phrase “persons in each state” is interpreted as applying to a state’s “inhabitants” – which is an “indeterminate” term that requires the president to exercise judgment.
The challengers painted a very different picture, arguing that the Constitution makes clear that everyone who lives in a state, whether they are in the country legally or not, should be included in the population count used to allocate seats in the House. The Constitution requires seats in the House to be allocated based on the “whole number” of “persons in each state,” they noted. People who are in the country illegally are “persons,” they reasoned, while the phrase “in a state” has long been understood to mean “usual residence” – that is, the place where someone lives and sleeps. The drafters of the Constitution included two exceptions to this general rule – for Native Americans and slaves – that confirm that everyone else who lives in the country should be included in the count for purposes of apportionment.
Federal laws also indicate that everyone who lives in a state should count toward its population for purposes of seats in the House, the challengers added. The Census Act requires the secretary of commerce to report “the tabulation of total population by States” to the president, who is then required to transmit “a statement showing the whole number of persons in each State” to Congress. “There is no reasonable way,” the challengers concluded, “to read ‘total population’ to exclude persons based on immigration status.” And indeed, the challengers observed, when the law was passed in 1929, the census had counted everyone in the country, including people who were here illegally.
Finally, the challengers told the justices that the Constitution and federal laws also require the census to be the only basis for the allocation of seats in the House. The Constitution indicates that apportionment should be based on the “numbers” determined by the census, while the Census Act outlines the steps for reapportionment based on the total-population numbers resulting from the census. Congress specifically chose to have the secretary of commerce and the president rely solely on the census for reapportionment to keep politics out of the process, the challengers argued. Using non-census “administrative records” – as the administration has proposed – to remove some noncitizens from the census’ total-population count would run afoul of Congress’ directive, the challengers said.
After Monday’s oral argument on, the justices are likely to issue their decision relatively quickly, to allow the reapportionment to proceed on schedule. But the tight timeframe for finalizing the census data and delivering the apportionment information raises another question outside of the courtroom: whether the administration, even with a favorable ruling, will be able to implement Trump’s memorandum as a practical matter. The New York Times reported last week that census officials have concluded that they will be unable to produce final population totals to be used in apportionment until Jan. 26. That would put the apportionment numbers under the control of the Biden administration, which almost certainly would revert to the policy of including all persons regardless of immigration status.
This post is also published on SCOTUSblog.