"Immigration and National Wages: Clarifying the Theory and the Empirics"

Gianmarco I.P. Ottaviano and Giovanni Peri on immigration and wages:

Immigration and National Wages: Clarifying the Theory and the Empirics, Gianmarco I.P. Ottaviano, Giovanni Peri. NBER WP No. 14188,
Issued in July 2008 [open link]: Abstract This paper estimates the effects of immigration on wages of native workers at the national U.S. level. Following Borjas (2003) we focus on national labor markets for workers of different skills and we enrich his methodology and refine previous estimates. We emphasize that a production function framework is needed to combine workers of different skills in order to evaluate the competition as well as cross-skill complementary effects of immigrants on wages. We also emphasize the importance (and estimate the value) of the elasticity of substitution between workers with at most a high school degree and those without one. Since the two groups turn out to be close substitutes, this strongly dilutes the effects of competition between immigrants and workers with no degree. We then estimate the substitutability between natives and immigrants and we find a small but significant degree of imperfect substitution which further decreases the competitive effect of immigrants. Finally, we account for the short run and long run adjustment of capital in response to immigration. Using our estimates and Census data we find that immigration (1990-2006) had small negative effects in the short run on native workers with no high school degree (-0.7%) and on average wages (-0.4%) while it had small positive effects on native workers with no high school degree (+0.3%) and on average native wages (+0.6%) in the long run. These results are perfectly in line with the estimated aggregate elasticities in the labor literature since Katz and Murphy (1992). We also find a wage effect of new immigrants on previous immigrants in the order of negative 6%.

Comments

[Real Person from the Real World]

The issue of LEGAL immigrants who are imported by foreign vendors for high paying jobs IS AN ISSUE. While some highschool grads may take some courses and luck into a good paying tech job, those with degrees in tech are competing out of school with some guy who was vetted into entry level jobs overseas, to compete with grads. Why should a company, who wants someone who can "hit the ground running" settle for someone they have to train when they can get someone supposedly with credentials for the job, who available right now, for contract.... a visa serf. And WHO is making money? The visa serf is swapped around by vendors who get themselves on as preferred vendors somewhere. The vendors agree on what the serf gets, usually far below the market, while they vend the serf at or just below market prices (by haggling over $1 or $2 on what they pay the guy among themselves). The only one making money is the foreign vendors. The visa guy keeps his mouth shut, smiles and works extra hours, so that if he last long enough, he gets a green card and free agent status. Then he can go out and mint money with the best of the American techies.

Let me explain in clearer terms: Vendor 1 has a client looking for a commodity skills IT guy for a 6 month contract. Vendor two has a visa serf on bench with those skills. Vendor 2 agrees to $55/hr (who knows what the serf is actually paid) and Vendor 1 talks him down to $54/hr and vends the guy at $85/hr or more to his client, whatever is the going rate, or just slightly below. Sometime vendor 2 gets another offer for $56/hr or $57/hr and pulls the visa serf out from under vendor 1, and the game starts over. The visa serf smiles, does what he is told by his visa owner, vendor 2, works hard, doesn't complain about extra hours, and if he survives long enough, he too can eventually command US rates as a free agent. Meanwhile, the US techie LOST job the job experience.

TELL ME, where are these figured in on this study?

[cursed]

This NBER study claims that as a result of immigration wages went up .6%. So here is an idea, lets double the amount of labor flooding this country and we can double the rate of wages. The intellectual paladins who compiled these lies did it for a reason. To get paid. What is your reason for referencing it?

[robertdfeinman]

We've had (heated) discussions about papers like this in the past. This one is representative of a main stream of theoretical economics work over the past few decades.

The idea is to take some data, postulate a model that describes it and then use this model to extract some correlations between variables of interest.

There is a fundamental problem with this approach, the results can be affected by the choice of data sets, their scope (time or group) and by the model chosen. With any finite set of data points it is possible to fit an infinite number of curves through the data. In most large-scale social science studies this isn't really an issue, because the number of data points is large enough so that uncertainties are kept to a reasonable level. This is the basis of polling, a few thousand individuals can closely approximate a population of millions.

The problem is more severe when the data set is small. This is the difficulty with drug studies where a patient group may only be 100-200 people. Results and dangers may be distorted.

In this study the population may be large (immigrants and unskilled workers) but the data set is actually fairly small since it represents aggregates over about a 30 year period. The total number of data points is about 100 or so. As a consequence the correlations claimed are unreliable.

I looked at similar data (by eye) and I concluded that it was the lack of union power that has caused the decline in the wages of unskilled workers. If you want to see my reasoning here it is:

Immigration Facts Debunked

I also claimed that NAFTA caused the collapse of the subsistence farming economy in Mexico which drove people north and led to the rise of immigrants ready to be exploited. Strong unions had no problem keeping people out of jobs when they wished to in the past, in fact this was a power that was often abused as the formation of the CIO showed and as issues about minorities in the building trades are still demonstrating until today. Lack of union strength allowed firms to easily substitute non-unionized cheaper immigrants. If the strong unions had still been existence the tide of immigrants would have had no effect, they wouldn't have been hired.

I don't think the economics profession is doing itself any favors allowing papers like this be published when they concern issues of significant public policy. Both the right and left can use the same tricks to support their agendas.

Perhaps we should stop using economic arguments and revert back to old-fashioned ideas of right and wrong and common morality. Doing right and being fair does not need to be supported by data. The golden rule doesn't say "do unto others.. as long as it is cost effective."

[2slugbaits]

I'm a supporter of increasing immigration, so I like the conclusions in the report, but I have some real heartburn about the methodology. The big problem is the simplifying assumption of a Cobb-Douglas production function. Why do I think this specification was selected more for the ease and convenience of the researcher than it was the appropriateness of the Cobb-Douglas model. Cobb-Douglas is good enough if you're doing a Solow kind of macro model or even if you're the BLS trying to calculate factor productivity, but it just doesn't cut it when you're trying to estimate elasticities of substitution between different qualities of labor inputs. A better choice would have been a translog model with capital, skilled labor, unskilled labor, energy and materials (five factors). A translog would have specifically modeled second order interaction terms between capital and skilled labor, capital and unskilled labor, energy and capital, etc. For example, somewhere I've got a BLS study that did something along these lines. My second beef is with the way the cross elasticities were reported as symmetrical. Why not use Morishima elasticities of substitution? A third problem is with the assumption of a constant elasticity of substitution (plus setting this at 1.00 as required by Cobb-Douglas). Contrary to what the paper claims, labor's share of income has not been quite as stable as the paper suggests. Over the 1990-2006 timeframe it drifted almost 5 percentage points. Fourth, the paper leans heavily on the total factor productivity term (TFP) because it sweeps a lot of questions under the rug. How much of that TFP is due to increases in technical skills of highly skilled workers? Fifth, since the Cobb-Douglas specification only used two inputs in the pre-nested model, the interpretation of the short run and long run elasticities isn't really correct. In a multi-input model the short run elasticity is identical to a two-input model because only one factor is changed and all others are fixed. So this effectively reduces to two inputs. The definition of the long run elasticity is that all inputs are allowed to change. This really doesn't have the same meaning in a two-input model. Sixth problem....oh forget it. I've gone on long enough.

[mik]

I'm a supporter of increasing immigration, so I like the conclusions in the report, but I have some real heartburn about the methodology.

Ends justify the means, eh?

You say you proved that the paper is garbage (I haven't read your reasoning, I place very little value in econ models), how can you like its conclusion? GIGO principle should apply.

And garbage is garbage, even if it looks appetizing.

[NoMoreBlatherDotCom]

I'll bet MattY wishes he hadn't linked to this.

[Peter Schaeffer]

2slugbaits,

Kudos. You actually read the paper (and understand the model).

[Peter Schaeffer]

This is sloppy economics. Even if you buy into the P/O model, the economics of low skill immigration are dismal. P/O don't even try to take into account the negative externalities associated with low-skill immigration. There are any number of issues.

1. Crime - Contrary to various claims, illegal immigrants are a very high crime group. Roughly 10% of the U.S. jail population is illegal (according to the GAO). Since illegals are only 4% of the population, they are a large net crime burden.

2. Taxes - The Heritage foundation has studied the tax costs illegal immigration. Each low skill immigrant household pays around $10,000 in taxes and consumes around $30,000 in public services. That's a net $20,000 loss to the American people. Here is an amazing fact. The average low-skill immigrant household consumes more in public services than it earns... (average earnings are below taxpayer funded services consumed).

3. Wages - If you actually read the P/O papers, you will find that they shows huge empirical wage losses for workers (native and immigrant) in California and elsewhere (correctly). However, these wages losses are simply a "magical" phenomena happening for some other reason. The P/O model also predicts that California should have higher than national wages (because of immigration) and natives should be moving to California to take advantage of the complementary workers (low-skill immigrants). Both conclusions are exactly wrong. Natives are fleeing California and wages are rock bottom (far below other states).

4. Contrary to P/O, immigrants are massively displacing American workers in the labor force. This is an empirical (reality based) conclusion. Take a look at Grogger, Hanson, and Borjas who wrote “Immigration and African-American Employment Opportunities: The Response of Wages, Employment, and Incarceration to Labor Supply Shocks”

The abstract tells the story

“The employment rate of black men, and particularly of low-skill black men, fell precipitously from 1960 to 2000. At the same time, the incarceration rate of black men rose markedly. This paper examines the relation between immigration and these trends in black employment and incarceration. Using data drawn from the 1960-2000 U.S. Censuses, we find a strong correlation between immigration, black wages, black employment rates, and black incarceration rates. As immigrants disproportionately increased the supply of workers in a particular skill group, the wage of black workers in that group fell, the employment rate declined, and the incarceration rate rose. Our analysis suggests that a 10-percent immigrant-induced increase in the supply of a particular skill group reduced the black wage by 4.0 percent, lowered the employment rate of black men by 3.5 percentage points, and increased the incarceration rate of blacks by almost a full percentage point.”

Now some of you may argue that these folks are on the right. However, some folks on the left have an even more devastating paper out. Take a look at Sum, McLaughlin, Khatiwada, and Palma who wrote “The Nation’s Temporary Guest Worker Program, the New Immigrant Workforce, and The Steep Deterioration in the Nation’s Youth Labor Markets”. A few quotes tell the story

“America’s teen and young adult labor markets have been devastated over the past seven years. Employment levels and rates of employment among all teens and most young adult subgroups (20-24 years old) have declined markedly since 2000, especially males, those with no post-secondary schooling, and youth from low income families”

“Declines in youth employment have been matched almost one for one with increased employment of new arrivals over the past 7 years”

5. Education – Unskilled immigration is damaging public education in the United States. This is a sad truth found in every study of America’s future labor force. Don’t believe me? See “Coming US challenge: a less literate workforce” (http://www.csmonitor.com/2007/0206/p02s01-legn.html). A useful quote

“US workers may be significantly less literate in 2030 than they are today. The reason: Most baby boomers will be retiring and a large wave of less-educated immigrants will be moving into the workforce. This downward shift in reading and math skills suggests a huge challenge for educators and policymakers in the future, according to a new report from the Educational Testing Service (ETS).”

The sad reality is that unskilled immigration is a major burden on the U.S. It must be stopped. At the end of the day we can have Open Borders or we can have the American Dream. We can not have both.