To What Extent Do Illegal Aliens Get Counted In The U.S. Census?

The 2020 census is moving towards its conclusion. A few days ago (July 21) President Trump took the occasion to weigh in with an Executive Order directing the Secretary of Commerce, when he reports the census results to the President, to provide sufficient information to enable the President to exclude from the “apportionment base” reported to Congress any “aliens who are not in a lawful immigration status under the Immigration and Nationality Act.”

Cue the usual outrage on the left. (E.g. from AlterNet.com, “Legal experts slam Trump’s ‘blatantly unconstitutional’ attempt to corrupt the census”; New York Times: “States Sue Trump Over Order to Exclude Undocumented Immigrants From U.S. House Seat Counts.”).

I suspect your reaction might be something different, like: What? Are you telling me that up until now they have been counting illegal aliens in the census for purposes of apportioning representation in Congress? How can that possibly be right? Do illegal aliens get representation in Congress even though they not only cannot vote, but also are not even here legally and are subject to deportation at any time?

Yes, that is exactly what the situation has been up to now. And by the way, it will probably continue to be the situation, maybe to a somewhat lesser degree, even in the face of the President’s Executive Order. By various estimates, there are probably about 4 Congressional seats at issue, most likely all of which go to Democrats currently and would go to other states and to Republicans if the illegals were excluded from the count.

Let’s go through the relevant background. First, there is the language of the Constitution. The applicable provision, which replaced language found in the original document, is found in the Fourteenth Amendment, Section 2:

Representatives shall be apportioned among the several states according to their respective numbers, counting the whole number of persons in each state, excluding Indians not taxed.

To implement the provision of the Fourteenth Amendment, Congress has enacted 13 U.S.C. Section 141, which contains the following language describing the task at hand:

The tabulation of total population by States under subsection (a) of this section as required for the apportionment of Representatives in Congress among the several States shall be completed within 9 months . . . .

So in one it’s “whole number of persons,” and in the other it’s the “total population.” But if you start thinking about it, you quickly realize that there are substantial categories of people whose inclusion or exclusion from “persons” or the “population” may not be obvious. For example, how about children? They can’t vote; so should areas with more children get more representation than areas with fewer children, even though the latter have more voters? You probably react that of course children should be counted — they are “persons.” And in fact, children are counted. Well, then, how about tourists who happen to be in the country on census day? And how about Americans who happen to be abroad on census day — should they be excluded? Americans abroad on census day in fact are counted, while tourists in the country in fact are not counted (or at least are not supposed to be counted) — which establishes the proposition that just because a “person” happens to be in the country on census day (even legally, in the case of a tourist) does not necessarily mean that counting that person is required or appropriate.

And then we come to the large and fraught category of immigrants. According to the most recent (2018) data I can find (from the Migration Policy Institute), there were a total of approximately 45 million “immigrants” (their word) residing in the United States in that year. This is not a small number, and constitutes almost 14% of total U.S. “population” as given by the census. (I put the word “population” there in quotes because the figure clearly includes illegals, but the issue here is whether illegals should or should not be included in “population,” at least for Congressional apportionment purposes.). Of the 45 million total immigrants, the MPI gives figures of 22.6 million as “naturalized U.S. citizens,” and 11.3 million as “unauthorized immigrants.” Presumably, the balance of approximately 11 million is what are called “lawful permanent residents,” generally holders of so-called green cards, but not citizens, and not entitled to vote.

With an average of 710,000 people in a Congressional district, the 11 million “unauthorized” immigrants potentially represent as many as 16 or so Congressional seats. However, note that if the illegals were distributed evenly throughout the country, their presence wouldn’t make any significant difference to apportionment. In the real world the “unauthorized” are concentrated in a small number of states. By MPI estimates, there are 3.1 million in California, 1.6 million in Texas, 0.9 million in New York, and 0.7 million in Florida, with just those four states therefore representing more than half of the total. Pew Research estimates that inclusion versus exclusion of the unauthorized in the census would affect three Congressional seats: one additional seat for each of California, Texas and Florida if illegals are included; one additional seat for each of Ohio, Alabama and Minnesota if they are excluded. Other estimates I have seen indicate that at least one more seat could be affected, going to California if illegals are included, and Illinois if they are excluded. You may look at this and think that the partisan balance would not be much affected, given that Texas and Florida are generally red states. But note that the illegals concentrate in areas of the Rio Grande valley in Texas that vote heavily Democrat; and in areas of South Florida (Miami, Fort Lauderdale) that also tend Democrat. It is highly likely that including illegals in the “population” swings at least 3 or 4 seats to the Democrats.

Take a look at the 2020 census form. You will note that it makes no effort whatsoever to determine immigration status of anyone who has come from abroad. The very first question is the key:

How many people were living or staying in this house, apartment, or mobile home on April 1, 2020? Here, you'll count everyone living and sleeping in your home most of the time, including young children, roommates, and friends and family members who are living with you, even temporarily.

Nothing there as to whether anybody who “is living or staying” in the residence is legally in the country. Go through the rest of the form if you will, and you will see that nothing addresses immigration status in any way. They explicitly ask everyone (including illegals) to respond and then make no effort to determine how many respondents are not authorized residents. Indeed, it’s the opposite: the Census Bureau makes every possible effort to obfuscate how many people that it is counting are unauthorized residents. From the MPI website:

Note: MPI is among a small number of organizations that generate estimates of the unauthorized population because the Census Bureau does not.

You may recall that at the beginning of the 2020 census process, the Commerce Department proposed adding a “citizenship question” to the census questionnaire. This then engendered a flurry of lawsuits, with multiple District Courts enjoining the question. The administration sought a stay in the courts of appeal, and ultimately in June 2019 the Supreme Court refused to grant the stay. Although the Court left it open for the administration to provide a better rationale for including the question, at that point, the administration gave up and the question was dropped.

I never understood the purpose of the “citizenship question,” and the administration was right to drop the fight. The problem is that a “citizenship question” does not go to the relevant distinction, which is not between citizens and non-citizens, but rather between the non-citizens who are legally here (about 11 million) and the non-citizens who are not legally here (about another 11 million). The non-citizens who are here as legal permanent residents clearly are part of the “population” and should appropriately be counted for purposes of apportionment under the statutory language. The case for counting for purposes of apportionment the non-citizens who are not authorized to be in the country is extremely weak in my view.

Anyway, we can now see from the July 21 Executive Order that the administration realized that it was going down the wrong path, and came up with an improved strategy. They figured out that the focus needs to be not on “non-citizens,” but rather on illegals.

But just because the new strategy is improved does not mean that it will easily carry the day. The President has ordered the Secretary of Commerce to provide information to enable the President to exclude unauthorized immigrants from the population totals used for purposes of apportionment. OK, but what is that information? Nothing in the responses to the census forms is going to provide that information. Presumably the Secretary will need to commission the Census Bureau to conduct some kind of a statistical survey to provide an answer as to how many people who responded to the census questionnaire are unauthorized residents, and as to which states they are located in. In the first instance, expect the Census bureaucracy to do everything in its power to obstruct the process and avoid producing results that would lessen Democratic power in Congress. Next, the results of any such survey that would shift power by even one seat will be vigorously challenged in court. Indeed, as noted above, numerous lawsuits have already been filed seeking to prevent the President from even finding out how many illegal aliens are currently being counted among the “population” in the census process.

But perhaps actually winning the battle of the census is not really the ultimate point. The ultimate point more likely is to force the Democrats into the open to admit that they are fighting for the ability to create multiple “illegal alien” Congressional seats, inhabited by few actual voters, to be represented in Congress by Democrats, while places like rural Alabama, Minnesota and Ohio get their representation taken away. That’s rather a hard position for the Democrats to defend politically.