Should The Census Count Illegal Immigrants?

The San Jose Mercury News ran a front-page article yesterday which was kind of interesting, as it posed the question: "Should illegal immigrants count in the census for determining how many seats each state gets in the House of Representatives?"

While at first glance, this seems like an easy question to answer, it really isn't. Historically, up until now, they have counted -- ever since the Fourteenth Amendment was passed.

First, the numbers. By their figures (credited to the "Connecticut State Data Center"), if the 2010 U.S. Census counts illegal immigrants towards apportioning House seats (as it has been doing), five states would lose House seats. Missouri, Illinois, and Michigan would each lose one seat; and Ohio and New York would both lose two seats. Three states would gain seats. Arizona and Texas would each gain two seats, and Florida would gain three. All other states would remain the same.

But if illegal immigrants were excluded from the census count, only four states would lose seats. New York, New Jersey, and Ohio would each lose one, and (surprise) California would lose two. Four states would gain seats, with one going to each of Montana, Arizona, and Texas. Florida would pick up two seats.

So it's an important question, especially for the eleven states involved. But is the answer obvious? Some history is needed to put the question in perspective.

[Fun census fact: An invention made for the 1890 census was a key milestone in the history of the development of the computer. See the history of the Hollerith Tabulating Machine and subsequent creation of IBM for details.]

From the original text of the Constitution (Article I, Section 2):

"Representatives and direct taxes shall be apportioned among the several states which may be included within this union, according to their respective numbers, which shall be determined by adding to the whole number of free persons, including those bound to service for a term of years, and excluding Indians not taxed, three fifths of all other Persons. The actual Enumeration shall be made within three years after the first meeting of the Congress of the United States, and within every subsequent term of ten years, in such manner as they shall by law direct. The number of Representatives shall not exceed one for every thirty thousand, but each state shall have at least one Representative..."

There are a few interesting things worth pointing out from the "original intent" of the framers of the Constitution. The first is the four classes of persons listed: "free persons" (free whites and free blacks); "those bound to service" (indentured servants); "Indians not taxed" (Indians who didn't pay U.S. taxes); and "other Persons" (slaves). Each was handled slightly differently. Free persons were all counted, of course, including those who could vote (white men owning property) and those who could not (everyone else, including women and free blacks). Indentured servants were likewise counted, even though they weren't free and couldn't vote. Indians were excluded, however, since they weren't even taxed, were not considered citizens in any way, and could not vote. Slaves, however, were counted (kind of). The "three-fifths" rule was a compromise between southern states (who wanted all the slaves counted, to boost their representation in Congress) and northern states (who didn't want them counted at all, to be "more fair" about representation in Congress). Nobody really liked this compromise, but it was deemed acceptable to all.

Now, basing a legal argument on this would be easy for either side of the question to do. If you were for counting illegal immigrants, you would point out that indentured servants and slaves (neither of which were citizens in any way, shape, or form) were indeed counted, in one form or another. There were lots of people who did not have full citizenship rights (such as the vote) who were nonetheless counted for the census' purpose. Therefore the intent was clearly to count all "persons" in the United States. Indians were considered (at the time) savages and therefore sub-human, therefore they weren't even "persons" (so the thinking went).

If you were against counting illegal immigrants, you would note that Indians didn't pay taxes and therefore didn't participate in the government in any way, and you would draw parallels to illegal immigrants living in the country today getting paid "under the table," tax-free. You might also note that slaves weren't counted fully, since they obviously were just a labor force imported into the country which nobody ever expected to become citizens, and draw similar parallels to today's situation.

In any case, it is no longer 1789, and things are a bit different in America today. For one thing, if we had followed the original ratio (one representative for every 30,000 persons) the House of Representatives would have 10,000 members today. Whew! Imagine getting anything done in such a huge group!

So how many persons does a representative represent today? Well, it depends. Using 2000 census figures [PDF download], the most populous state in the country (California) has one representative for every 639,088 persons. The least populous state in the country (Wyoming) has one representative for every 493,782 persons. But the largest state population represented by only one congressional district (Montana) has one representative for 902,195 persons. As you can see, representation fluctuates wildly between less than 500,000 per House member to over 900,000. So even with the way it is figured now, there is a fairly wide range of how much your vote is diluted, depending on where you live. Sadly, the Mercury News article ignores this fact, and while doing so, gets the ratio backwards:

With the nation's population of unauthorized immigrants growing by about 500,000 a year, the report says illegal immigration is concentrating the power of voters in states such as California, Texas and Arizona, which have more seats in Congress per legal resident than many states where the number of illegal immigrants is much smaller.

Sure, illegal immigration is "concentrating the power" of California voters, but assuming that two House seats' worth of illegal immigrants lives in California would still give a figure of 664,150 voters per House seat -- still comfortably in the middle of the range of all U.S. states. And measuring "seats in Congress per legal resident" would be like measuring gallons per mile for a vehicle -- it's completely backwards -- and should read "legal residents per seat in Congress."

The article goes on to state (without offering any context):

The dilution of the principle that every voter has an equal voice in the nation's political life is one ramification of the immigration issue, say Rodriguez and immigration experts. [Orlando Rodriguez is the demographer who wrote the report the article is based on.]

But, as I have stated, people may think there's a "principle that every voter has an equal voice," but in reality there is wide disparity in this "equal voice," with a Wyoming voter currently having almost twice the "voice" that his neighbor to the north in Montana has. This is not mentioned anywhere in the article.

Politically, the effect of counting immigrants is a tough one to figure. If you count immigrants in the census, some solid-blue Democratic states would lose a total of four House members ("Reps"), as well as three Reps from some mixed (or "purple") states. Texas, a solidly red Republican state would gain two Reps, but two fairly evenly purple states would gain five Reps as well.

If you don't count immigrants, three solidly blue states would lose the same four Reps, as well as one Rep from one purple state. Two red states would gain the same two Reps, and the same two purple states would gain three Reps.

In both cases, red states are up two Reps, blue states are down four Reps, and the rest migrate between various different purple states (of differing shades of purple). So it's not easy to see what the political benefit truly is to either side. What is clear is that there is a net migration of population from the Northeast and Rust Belt (or Midwest) states to Florida, Texas, and Arizona, and both scenarios highlight this to varying degrees. It may turn out to be more of a regional issue than a party issue.

That's assuming, of course, that demagoguery and scapegoating don't rear their ugly heads. Which is probably too much to ask, given the propensity of Republicans to shoot themselves in the foot (for the entire next generation) by demonizing immigrants.

[Not-so-fun census fact: When the order was given to intern the Japanese Americans on the West Coast, the Census Bureau voluntarily offered up data on where such persons lived, including street addresses. This completely violated every privacy rule the Census Bureau had ever had. Think about that while you fill out your next census form.]

But this has a better chance to sway middle-of-the-road Americans than most immigration issues (like the insane idea of "let's just ship them all home tomorrow"), because it would be pitched as an issue of "fairness." The argument will go: Illegal immigrants shouldn't count, and it's unfair to everyone else to count them. The problem is, the way the law is currently written is pretty unequivocal. It comes from 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."

"The whole number of persons" is pretty straightforward. It does not say "citizens" or even "residents" but just flat out "persons in each state." Which means that any change would have to at least pass Congress, if not actually be an Amendment itself. Reportedly, one congresswoman from Michigan has introduced a bill to do so. Which means the fight over how to interpret the 2010 census has just begun.

Maybe we'll wind up counting illegal immigrants (if it were even possible for the census to enumerate them accurately -- which is a might big "if") as the framers of the Constitution may have intended: as three-fifths of a person. Much like the first time it was written into the Constitution, this would be a compromise almost nobody would like, over how to count the labor class in American society. Maybe it's a fitting end for this debate, so that two hundred years from now students will look back and say: "How did they come up with such a stupid idea?"


Chris Weigant blogs at: