Ottoviano-Peri and the CEA vs. George Borjas on the Distributional Consequences of Immigration

George Borjas appears to get the economic theory not quite right:

The Borjas Blog: An Inconvenient Truth That Somehow Didn't Make The CEA Report: As I mentioned in a previous post, the CEA seems to have concluded that if one allows for complementarities between immigrants and natives [i.e., that unskilled native-born workers can serve as translators and team bosses for their coworkers who don't speak Englidh]... the long-run gains from immigration are somewhere between $30 to $80 billion per year. A careful reading, however, indicates that the CEA doesn't quite say that....

As I suspected, the $80 billion number does not mean what most people would probably take it to mean. Economic theory predicts that the long-run gains from immigration to the pre-existing population must be zero—-even when there are complementarities between immigrants and natives and even if those complementarities are incredibly strong. In the long run, capital adjusts fully until firms wither away all the excess profits from the initial wage depression...

Now that's simply wrong: "capital adjusts fully" means that more future investments are made in high-productivity areas to which migrants move and fewer in low-productivity areas from which migrants came. Returns on savings are thus higher--and because the pre-existing population are savers, they benefit. So do the migrants. The long-run model in which Borjas is implicitly working is one in which real wages for non-movers are constant by assumption: so there are no losers, only gainers.

Borjas then shifts to the Ottaviano-Peri model:

The CEA used the Ottaviano-Peri result that the complementarities helped natives and calculated how much natives gained as a result. This is what they say: "Multiplying the average percentage gains by the total wages of US natives suggests that annual wage gains from immigration are between $30 billion and $80 billion." But they completely ignored the fact that the same complementarities that supposedly help natives also hurt [previous] immigrants, and by quite a bit. In other words, the CEA uses a strange definition of who “we” are: including only native-born workers and ignoring the millions of immigrants already here, who are affected by yet more immigrants.... Had the CEA taken the immigrant losses into account, the Bush administration would have had to report that the net gains from immigration for the pre-existing population are equal to... ZERO!

This is true only if you believe that previous immigrants are exactly like brand-new immigrants--that they don't become more like native-born over time as they gain social knowledge and English proficiency.

And there's actually more embarrassing news, for a theorem is a powerful thing.... [T]he average [long-run] wage change [note: not the real income change] in the [pre-existing] population must be zero.... If there were 15 million [previous] immigrant workers, each immigrant worker must lose $3,333 annually--and the $50 billion gain accruing to natives must be entirely offset by the $50 billion loss accruing to immigrants.... Imagine the headlines had the CEA reported that immigration during the 1990s led to a $3,333 drop in the average earnings of pre-existing immigrants! This is not the spin the White House was looking for, but it is a direct implication of the spin they did put out. What an inconvenient truth! I wonder if the compassionate conservatives will shed a tear about the huge wage losses suffered by pre-existing immigrants.

If I had Giovanni Peri here at hand, I think that he would say that increased immigration is very good for new migrants, good for savers worldwide, good for native-born workers, good for previous immigrants who have substantially assimilated--social knowledge, English proficiency, et cetera--and probably bad for previous immigrants who have no assimilated. And he would also say that the model goes haywire and is untrustworthy when the number of non-assimilated immigrants is small, and that that going haywire is where the very large income losses for previous migrants is coming from.

And, of course, the thing to object to in the turn this entire debate has taken has been the failure to focus evenly on the consequences for all stakeholders in global migration--look at what happens to everyone, not just one particular group that is convenient for your current political position.

But I will ask him.

E.P. Lazear et al. (2007), "Immigration's Economic Impact" http://www.whitehouse.gov/cea/cea_immigration_062007.html

G. Ottaviano and G. Peri, "Rethinking the Effects of Immigration on Wages," NBER Working Paper 12497 (2006)

Comments

[save_the_rustbelt]

I'ma Republican and I don't believe anything that comes from the White House, even if it is from fast Eddie Lazear.

[Tyrone Slothrop]

"the thing to object to in the turn this entire debate has taken has been the failure to focus evenly on the consequences for all stakeholders in global migration--look at what happens to everyone, not just one particular group that is convenient for your current political position"

That's not necessarily fair -- the question of which group to concern yourself with in this debate *is* a political question in the classic, non-pejorative sense. Which is not say that I disagree with where you're going with this, but deciding that a nation's economic policy ought to advance the interests of its citizens before those of others is a principled position, if disagreeable, and not necessarily one of convenience.

[Tuco]

"look at what happens to everyone, not just one particular group that is convenient for your current political position."

From the point of view of what happens to everyone the discussion needs to expand to include the degradation of environment, destruction of habitat, health issues from increased pollution (affecting primarily children and the elderly - how sweet is that?) and their economic costs to society, quality of life (from increasing gridlock around an increasing number of cities), and competition for limited resources, to name a few. Over the past fifty years America has seen population growth equal to a new Houston every year (and yes that's primarily from immigration). That's fifty Houston's in fifty years. (In case you were wondering Houston is America's fourth largest city behind NY, LA and Chicago.) How long can we keep adding a new Houston every year before we finally notice how destructive unchecked immigration is? This needs to be more than an academic/economic/political issue if it is to reach the level of a Grasping-Reality-With-Both-Hands issue.

America - 9,629,091 square miles, 300,000,000 people

Australia - 7,686,850 square miles, 20,500,000 people

[Ponzi Q. Globalization]

"...deciding that a nation's economic policy ought to advance the interests of its citizens before those of others is a principled position..."

Tyrone, I hope you realize that saying this makes you a racist in the eyes of many neoliberal globalizationoids. Be strong.

[Matt]

Borjas is a serial book-cooker on immigration. It's always wise to read him very carefully and check everything he says.

[randomeconstudent]

"America - 9,629,091 square miles, 300,000,000 people

Australia - 7,686,850 square miles, 20,500,000 people"

Japan- 144,859 square miles, 127,433,494 people

Further, Japan is only the 30th most dense country in the world, while the US ranks 172nd (Australia is one of the ten least densely populated countries in the world.) We have a long way to go before reaching a level where population density in the US is a real problem, and the environmental impact of immigration should net zero (more pressure on the US environment, less on the Mexican environment.)

[Sonia]

Tuco states: "Over the past fifty years America has seen population growth equal to a new Houston every year (and yes that's primarily from immigration)." From the US Census US population growth between 1960 and 2000 was approximately 100 million (to 281 million). Foreign born population increase from about 10 million to about 31 million over the same period or roughly 20% of the total growth. Between 2000 and 2005 about 50% of the the total population growth of 10 million is accounted for by the growth in the foreign born (Pew Center). So only in the last 5 years does foreign born population growth account for 50% of the growth.

Of course some areas of the country have experienced much higher immigrant growth in the last 5-10 years. In these areas there are clearly many domestic winners and losers, witness the divided Congresional delegations from Arizona and New Mexico.

The economics debate here re complementarity and capital flows has not yet been addressed by the comments. Rather than disbelieving CEA, I suggest viewing the analysis as 1/2 the story. I would like to see some more analysis of the theory.

[Matt]

If we pack immigrants into energy efficient cities, then they can, hopefully, provide intermediate specialization that improves overall efficiency.

The U.S. has more efficient infrastructure, more land.

The problem is, why doesn't a sufficiently large N cover all the people permeatations that lead to optimum efficiency?

In other words, if I examine LA, and decide some intermediate people packing can produce overall efficiency gains, then do I not reach a point in which the cost of packing itself outweighs the small efficiency gain? I mean, the cost of increasin N outweighs the gain that N+1 provides?

[notsneaky]

Brad's right overall but I think it's important to spell out the effects on various groups in more details.
Suppose for simplicity that there's native workers and two waves of migrants, one in say, 1990 and another in 2000. Ignoring capital owners, the 3 groups of stake holders are natives, first-wave migrants, and second-wave migrants.
Suppose further that when they first arrive immigrants are complimentary to native workers a la OP, but that after ten years a fraction of them (say p) becomes "assimilated" and becomes a substitute for native workers. The (1-p) share of first wave workers who do not assimilate are perfect substitutes for the second wave migrants.

(Note there's an implicit economic assumption here that assimilation has benefits, otherwise, why do it? Why not stay complimentary?)

As a result the pattern of gains and losses for each group over time is:

Native workers =
1. Gain initially with an influx of complimentary first-wave migrants.
2a. Gain again with a further influx of complimentary second-wave migrants, but
2b. Some of these gains are offset by an increase in perfectly substitutable labor (for them) due to assimilation of first wave migrants.

First-wave migrants =
1. Gain big as they move from unproductive employment back home into much more productive employment in US.
2a. Those who assimilate make further gains as there are benefits to assimilation (see note above).
2b. Those who do not assimilate see some loss due to competition with second wave migrants who are perfectly substitutable for them. BUT THIS LOSS IS RELATIVE TO THEIR POSITION IN 1) NOT RELATIVE TO THEIR POSITION BACK IN THEIR NATIVE COUNTRY. It's not crazy to think (as in "it's crazy to think the opposite") that the gains in 1) are way way more than the losses in 2b.

Second wave migrants =
1. Gain big as they move from unproductive employment back home into much more productive employment in US.

[Chris]

Can you imagine Lazear writing another article like the one below now that Bush is his boss?

http://www.aeaweb.org/annual_mtg_papers/2006/0108_1015_0304.pdf

[Neil B.]

A lot of that argument about the supposed benefits of immigration (extra work for the rest of us) is a variation on the broken-window fallacy, isn't it?

[notsneaky]

Peter,
I think there's a bit off ... sleight of hand, going on. Borjas is right that with international capital mobility, etc., there's no permanent benefit to savers - contra Brad. But overall this seems like a minor point to argue about.
Borjas' original post was mostly about how if natives and new migrants benefit, then, in the long run, it must be that old migrants that lose. I think the major points that Brad was making were that 1) if some of the old migrants become more similar to natives (ie complimentary to new migrants) then the 3,333$ loss to old migrants that Borjs quotes is no longer true - it would be much less and 2) the "loss" to old migrants isn't really a loss if your baseline of comparison is the incomes they would make back home in the less productive economy.

Also, related to something you said at MR - in Ottaviano/Peri, it is NOT assumed that natives and immigrants are complimentary. the elasticity of substitution/complimentarity is ESTIMATED from the data.

Neil B.,
No, it's about wages.