Illegal Aliens 101: How Many Illegals Are There?

I've received a number of emails asking me to provide some basic information about immigration in the United States. It's a great idea and I'm happy to satisfy the demand.

So let me start with a central issue in the current debate: How many illegal immigrants are there? In recent days, I've seen numbers ranging from 6 to 20 million being floated around--depending on the spin that is being emphasized in any given article.

The basic approach used to calculate this number was first formulated by Jeffrey Passel and Robert Warren (who was then, if I recall correctly, the chief statistician at the INS) and was published in Demography in 1987.

The idea is this: The U.S. Census Bureau regularly surveys the population (as with the decennial Census), and these surveys provide a count of how many foreign-born people live in the country. This count will include many illegals as they try very hard to count people "living in the shadows."

At the same time, the immigration bureaucracy keeps good records on how many permanent residence visas they've granted (i.e., how many "green cards" have been given out). By making a few assumptions about the mortality rate of immigrants, about the number of foreign students in the country, etc., we can predict how many foreign-born people should be living permanently in the country.

The number of illegal immigrants is given by the difference between the Census count of foreign-born persons and the number of foreign-born persons who should be in the United States.

The Department of Homeland Security's website contains reports that summarize the results of this exercise for 2000 and 2005. And this is what the official counts say: The 2000 report estimates 7 million illegal immigrants as of January 2000. The 2005 report estimates 10.5 million illegals as of January 2005. Further, the 2005 report claims that the number of illegals was growing at the rate of 400,000 annually. So if this rate of growth has been roughly the same over the past couple of years, we are nearing the 12 million mark.

Are these numbers credible? Yes....but. There's one underlying assumption in the calculation that, I think, probably leads to an underestimate of the number of illegals. No matter how hard the Census Bureau tries, it cannot possibly count every illegal immigrant in the country. There are some shadows that are way too dark even for the Census Bureau to penetrate. So one has to make an assumption about the rate of undercount of illegals. Before I tell you what they assumed, let me ask you to just think about it for a moment: what fraction of illegals do you think are missed by the Census?

The DHS assumption is that the rate of undercount is 10 percent (see point f on page 2 of the 2005 report). Maybe it's just me, but I think that assumption is on the low side. Given that, my hunch (and it's only a hunch as I have no idea about what the true rate of undercount is) would be that the "true" number of illegal immigrants in the United States right now could easily be 13 to 14 million.

(An interesting aside: For reasons unknown to me--although conspiracy theorists could have a field day--the count of the number of illegal immigrants in 2000, as reported by the contemporaneous 2000 report, is 7 million. But the number of illegal immigrants in 2000 as reported by the after-the-fact 2005 report, is 8.5 million, making it seem as if fewer illegals entered during the Bush years).

Comments

[Lucius Vorenus]

All of the numbers from the Census Bureau are politicized - none of them can be trusted.

If The Blob thinks it advantageous to underestimate the illegal population, then it will underestimate the illegal population [and, in an alternate reality, where Vulcans are green & evil, and behave like renegade Romulans, The Blob will overestimate the illegal population].

By contrast, one of the more interesting attempts to quantify the illegal presence is to count the number of duplicate FICA payments to the SSA [I believe that among the popular commentators, Neil Boortz is the one who has really been hammering home the importance of this approach].

The numbers I've heard are something like 9M duplicate payments on identical social security numbers, which would indicate AT LEAST 9M illegals in the "legitimate" economy.

I googled a little, and found this Judicial Watch article, which quotes 10M TIN's issued, and almost 8M known bad W2's:

http://www.corruptionchronicles.com/2007/04/irs_caters_to_illegal_immigran.html

Then the interesting question becomes: What percentage of illegals work in the legitimate economy, and what percentage work in the underground, cash-only economy [which makes no payments to the SSA - recall the failed Zoe Baird nomination].

Finally you ask yourself: Okay, what percentage of "illegals" are minor children? [And we'll throw their anchor babies into the pot, the legality of illegal anchor baby-ism still being largely undetermined at this point]. And that, in turn, begs the question: What percentage of adult illegals "work" in the first place? Surely some of them must stay at home to watch the kids - they can't all be in state-subsidized day care, can they?

Anyway, depending upon your assumptions, you could come up with all sorts of figures.

For instance:

1) Assume 9M adult illegals in the "legitimate" economy.

2) Assume 50% of adult illegals work the legitimate economy, and the other 50% either work the underground/cash-only economy, or else stay at home with the kids.

3) Assume a 1-1 ratio between adult illegals and their minor children.

Then that gets you to

(2 X [9M + 9M]) = 36M

which probably isn't that far off the actual number.

[Patrick R. Sullivan]

'...making it seem as if fewer illegals entered during the Bush years'

Almost certainly true since after 9/11 it has been much harder for anyone to enter illegally.

[Lucius Vorenus]

EDIT: "10M TIN's" should read "almost 11M TIN's".

Patrick R. Sullivan: Almost certainly true since after 9/11 it has been much harder for anyone to enter illegally.

You're being sarcastic, right?

[Steve Sailer]

The number of Hispanics found in the 2000 Census was about 10% higher than the Census Bureau was saying the Hispanic population was, so

- The Census did a better job of finding illegal aliens in 2000 than in 1990.

- We still have no idea how good a job it did.

[jason voorhees]

Are there ways of calculating US immigration levels using Mexican and South American data sources? For instance, what if people living in the shadows here were not living in the shadows there? Then it might be possible to construct estimates here if we knew more about differences over time there. Or something like that. Just wondering.

[Lucius Vorenus]

jason voorhees: Are there ways of calculating US immigration levels using Mexican and South American data sources? For instance, what if people living in the shadows here were not living in the shadows there? Then it might be possible to construct estimates here if we knew more about differences over time there. Or something like that. Just wondering.

Well, you could try to figure out what the Central American populations "should" have been relative to what they appear to be [although I imagine that Latin American census numbers are even more highly politicized and more worthless than American census numbers, if that's possible].

Anyway, along those lines, there's been a flurry of articles with the following theme:

For Mexico, emigration to U.S. surpasses deaths.
http://www.google.com/search?safe=off&q=For+Mexico+emigration+to+US+surpasses+deaths

See also e.g.

Emigration depresses Mexico's countryside
http://www.freerepublic.com/focus/f-news/1603535/posts

[Lucius Vorenus]

There's some new chatter today about the question of known fraudulent data at the Social Security Administration & the IRS.

See e.g.

Congress Should Get the Social Security Administration to Name Names
http://corner.nationalreview.com/post/?q=NzVkNGYyYjZmOGE2ZDlhNGM3YjViOTViMjQwZTJkODg=

and the older article

Demand-Side Immigration Reform
By Terence Jeffrey
http://www.townhall.com/columnists/column.aspx?UrlTitle=demand-side_immigration_reform&ns=TerenceJeffrey&dt=05/23/2007&page=full&comments=true