Thursday, May 05, 2016

Undercounting illegals overstates purchasing power parity

It feels like I must have glossed over it before, but flipping through the annotations in Ann Coulter's ¡Adios America! I'm not finding any references to it. I figure that if I'd find it anywhere, it'd be there.

The seemingly obvious thought just occurred to me that if the number of illegal immigrants in the US is significantly underestimated by the Census' Current Population Survey, to the tune of several million--with some estimates asserting that it is underestimated by nearly 20 million people--then the purchasing power parity (PPP) figures for United States will be correspondingly overstated.

Per capita purchasing power parity--a measure of monetary standard-of-living that is designed to allow for comparisons to be made between different countries--is calculated by dividing GDP as measured in purchasing power parity by population size. The CIA factbook puts US PPP at $56,300 for 2015. If there are 20 million unaccounted for illegals here, however, that figure drops to $52,640.

There is, then, an obvious incentive for the federal government to low-ball estimates for the number of illegal immigrants--and unaccounted for people more generally--in the country to make income and purchasing power statistics look better than they actually are.

Am I missing something here?


akarlin said...

You're not missing anything.

In fact, this would be one possible (partial) solution of the "paradox" that the US is far richer than its average IQ would tend to warrant.

Ben Kurtz said...

What about the numerator? That is, the estimate of PPP GDP? Haven't looked up how that number is built, but if it relies on estimates of economic activity that use population as an input variable then maybe that goes up as well somewhat?

Also, about the PPP adjustment - to the extent that might rely on average labor costs as an input, would having more illegal aliens than expected cause the average wage to be lower than reported (on the assumption that illegals tend to work for rock-bottom wages), and lower average labor costs gives you a better PPP adjustment multiplier and therefore a larger numerator?

Not sure that these would have any effect at all, let alone whether such effect would be larger or smaller than the effect of enlarging the denominator like you propose - but could you lift the lid on these calculations and hazard a guess?

Mil-Tech Bard said...

>>Am I missing something here?

No you are not.

Anonymous said...

The incentive is to overcount illegal immigrants, because they tend to cluster in more liberal areas of the country. Congressional apportionment, formulas for distribution of federal funds and grants, and other federal goodies, are based on total population, not population of citizens. Therefore, having more illegals in your state gives you more votes in congress, more fedral $, more federal workers, etc.

Audacious Epigone said...


Great insight, thanks. It could also inflate productivity figures, which are perpetually used in the service of arguing for more immigration--"They can't be productive in Mexico, but here they can be. We'll be richer with them here and so will the world as a whole!"


PPP GDP is just GDP with currency value adjustments, so I don't think it would be effected. GDP isn't measured by population.

Mil-Tech Bard,



Census figures are federally administered. Maybe by puffing up blue states the federal government puffs itself up, but that seems like a populist headache that wouldn't be worth it.

The Z Blog said...

According to the GAO, $25 billion a year is sent from the US to Mexico by individuals. These are all small amounts so a little math suggests the number of Mexican nationals in the US is significantly higher than the government claims. There was a study about ten years ago by Bear Stearns using remittance data that pegged the illegal population at 20 million.

In the more rugged neighborhoods like where I live, there's a whole slab of the local economy that caters to the needs of illegals. Above ground you have prepaid cell phones, wire transfer services, etc. below ground you fake documents, social security numbers, etc. You also get used to seeing bank machines that allow illegals to deposit cash and wire it home.

My bet is the illegal population is at least 25 million at any one time.

tanabear said...

I've wondered about this as well. Wouldn't this also imply that poverty rates in the US are much higher than reported?

Audacious Epigone said...


Moneygram and Western Union probably have additional relevant data.


Income figures are assessed by representative surveying so undercounting the size of the total population wouldn't directly cause underreporting, but since illegals are mostly impoverished, yes, in that sense it would.

Dan said...

Total energy consumption has been more or less flat.

One chart says growth says the compound growth of energy usage was negative from 2001 to 2011.

I feel like enormous numbers of uncounted illegals would show up there.

Total vehicle miles travelled seems not to be skyrocketing.

Vehicle sales:

Mil-Tech Bard said...


The flat growth in energy consumption is being attributed to the high cost of energy, yet when prices came down recently the usual jump in American energy consumption did not come back to anywhere near the degree it did so in the past.

What the price drop has unmasked is that there is an enormous and growing trend seeing the replacement of the transportation of 'things' for the transportation of information.

This is part of the reason that "Big box" retailers like Walmart are seeing a shrinkage in their retail sales.

When you can download a e-book in lieu of the dead tree edition on the scale we are currently seeing. Retail big box stores specializing in selling dead tree editions of books suffer. See the fate of "Borders Books."

Walmart's sales drop is showing that this trend has accelerated and entered into some of Walmart's core retail volume niches.

Audacious Epigone said...


Unless, of course, the US population is getting concurrently poorer over the same period of time.

But even if the undercounting has been systematic, steady, and perpetual, it'd still mean the average US resident is poorer than the official stats make him appear to be.