As folks who have followed this website for a long time know, the decennial census is something that I consider to be a very big deal.  And, while perhaps not as detailed as we did it back in 2011, I am hoping that we will do something as the numbers come out in the spring about what the numbers might mean for our chances at keeping and increasing the Democratic majority in the House.

Redistricting involves two action at the federal level and at the state level.  At the federal level, the results of the census are use to determine how many representatives each state gets (often referred to as apportionment).  At the state level, assuming that a state has more than one representatives, redistricting involves drawing the lines so that each district has roughly the same population (no more than a 5% gap between the largest and smallest district and preferably smaller).   At the current time, of course, we are dealing with actions at the federal level.  The ball only shifts to the state level once apportionment has occurred and the Census Bureau has released the detailed count (breaking population down to census blocks) to each individual state on a rolling basis.

The federal part of the process comes first and involves two steps:  one involving raw data and the other involving the application of a formula to that data.    The first step is the census finalizing its state level population numbers. According to federal law, by January 1, the Census Bureau is supposed to report its numbers to Secretary of Commerce who is to forward those numbers to the president.  Upon receipt of those numbers, the President is to calculate the number of representatives that each state is entitled to and, by January 10,  forward a statement setting forth the population of each state and the number of representatives that each state will have in the next Congress.  The calculation is done by the “method of equal proportions” (one of several mathematical formulas used to “fairly” allocate partial seats).

For those mathematically inclined, this formula is used to calculate which state is entitled to the next seat.   Since the first fifty seats are assigned one per state as a matter of the constitutional requirement that each state gets one representative, it is used to calculate a list from the 51st seat to the 435th seat.  Of course, it is possible to also use it to calculate what would happen if two additional states (D.C. and Puerto Rico were added) or what would happen if you expanded the House to 449 members.  The basic formula is that for each seat you calculate an adjusted population per seat number.  This number is the total population of the square root of the current number of representatives for the state times that number plus one.  Thus, for the fifty-first seat, you would divide each state’s population by the square root of two (the current number of seats — one — times the next seat (one plus one, in other words, two).  This divisor will always be somewhere between the current number of seats and the next seat.  For obvious reasons, California as the largest state will get seat number 51 (as all states are divided by the same divisor for that seat).  But as each state gets each additional seat, its adjusted number continues to decline and it finds itself moved down the priority list. Thus, using the 2010 Census numbers, California was the first state to get a second seat followed by Texas, but California got a third seat before New York or Florida got a second seat.

Now, normally, this process is not that controversial.  There is always some quibbling between the state at 435 and 436 on the list over the accuracy of the count, but most of those disputes go nowhere.  But, of course, the Orange Menace does not like to follow the rules if they do not benefit him.  Earlier this year, President Trump — encouraged by discussion among the anti-immigrant crowd — decided to adopt one of their ideas.  For the past decade, some parts of the Republican Party have been proposing that states do not need to consider unlawful immigrants when it comes to redistricting.  According to this theory, the only number that matters is voting age citizens.  This wacky concept is based on the shorthand that lawyers use to describe the rule governing redistricting — “One man, one vote” or if you want to update the quaint language from the original cases from the 1960s “One person, one vote.”  Based on this theory, Trump directed the Commerce Department to exclude illegal immigrants from the numbers that it used to calculate reapportionment to the extent possible.  Of course, the benefit to Republicans is that a lot of immigrants live in blue states and urban areas that tend to vote for Democrats.  Excluding immigrants from the count would reduce the total number of representatives from blue states (and increase them in red states) and, within each state, would reduce the number of representatives from urban areas which vote for Democrats and increase the number of representatives from rural areas which vote for Republicans.

There were two problems with Trump’s memorandum — one legal, one practical.  On the legal side, the memorandum is blatantly unconstitutional.  Article I and the Fourteenth Amendment make it clear that — excluding Native Americans not subject to taxes — the Census is to count (and seats in Congress are to be apportioned based on) the whole number of persons in the state.  The Constitution makes a distinction in various spots between citizens and persons; so there is no real argument based on past practice, plain language, and the intent of the framers that any president could exclude legal immigrants from the count.  On the other hand, the small number of cases dealing with the census over the years has made it clear that the count is based the location where a person is an inhabitant (i.e. where she regularly lives).  Thus, if you are on a business trip on April 1, you are still counted at your regular residence.  On the other hand, college students and prisoners get counted at their current location even if they may intend to live elsewhere when the school year or their sentence is done.  While there may be some illegal immigrants who do not qualify as inhabitants of the United States, it is clear that many of them would.  In addition to the plain language of the Constitution, many of the statutes governing the census also refer to persons rather than citizens.

And this leads to the practical problem, there is no database of all illegal immigrants.  And the Supreme Court precluded Trump from asking a question about immigration status on the Census (which is part of what led to the current directive).  There are some partial databases (those being detained at the border or with pending removal/deportation cases) which might allow some adjustments to be made.  But the completeness and accuracy of those databases and the actual number on them is a matter of debate.

For the first reason, states with a high immigrant population and interest groups representing immigrants filed a case challenging Trump’s directive.  And the lower courts found for the states and interest group.  But the Supreme Court opted to take the case on an expedited basis.  But the second reason became crucial as the Supreme Court addressed the case.

At the present time, with a mere two weeks to go before the deadline for the Census Bureau to wrap things up, nobody knows the numbers that might be excluded under President Trump’s order.  And, it appears that there are enough other problems with the slipshod way that the Trump Administration handled a census during the COVID-19 crisis to cause the Census Bureau to miss its deadline — maybe long enough that it will be President Biden that gets the numbers from the Census Bureau (with a high likelihood that Trump’s memorandum will be seat aside at 12:01 p.m. on January 20).

Facing this reality (and not wanting to decide the ultimate issue of who could be excluded from the count), the conservative majority on the U.S. Supreme Court issued an unsigned opinion today vacating the lower court decision by a 6-3 vote with directions to dismiss the case “without prejudice” (meaning that it can be refiled later).  The theory underlying the majority’s position is that, given the indefinite impact of Trump’s attempt to violate the law, we don’t know which states will be harmed.  If we are talking about removing 10 million people from the final census count, there is a good chance that such an adjustment could cost large states a seat or two.  But if we are talking about removing 100,000 people from the final census count, there might be no impact.  The dissent — authored by Justice Breyer — on the other hand, thought that the risk of damage was substantial enough to allow the states that might be harmed to sue now to prevent any possible harm before we reach a more critical stage of the process.  And the dissenters (Justice Breyer, Justice Kagan, and Justice Sotomayor) thought that it was clear that the policy is illegal under the statutes governing the census regardless of its ultimate impact.

The effect of this decision is a punt.  And it is a misguided punt.  We are less than two weeks from the deadline for submitting the Census Bureau report.  While it is likely that the deadline will be missed, we will know in just over a month rather the memorandum is being implemented.  If the duty of reporting the apportionment numbers falls to President Biden, then the case becomes “moot” (i.e. the dispute disappears) as President Biden will cancel the memorandum.  But, if President Trump gets to act, we will know what the harm from the memorandum is.  Since the issue of standing and ripeness will be clear at that point, holding this case would do no harm and would have left the case in a position where the Supreme Court could  have quickly resolved the merits if that become necessary.  Requiring this case to restart in three weeks at the trial court is a waste of judicial resources and there was no need for the Supreme Court to rush to get this case out.  And the issuance of an opinion during the recess between argument sessions is unusual and tends to be limited to emergency opinions.  For the very reasons cited by the majority, there was no emergency.

Despite the desire of the conservatives to avoid addressing the merits, it appears likely that Trump’s attempts to cook the books will fail due to his own failings as an administrator and executive.  So with any luck, the next time that we are here to discuss the census will be in February 2020 when we will have honest numbers.  Of course, due to the errors of the past four years, the accuracy of the numbers may leave a lot to be desired.