The Alas Debate Annex

[Crossposted on "TADA" and on "Alas"]

Felix Salmon writes:

Never mind the stimulus vs austerity debate: here’s something that both sides should be able to get behind. It’s a simple legislative fix which increases tax revenues without raising taxes; which increases the demand for housing; which increases the economy’s productive capacity; and which boosts wages for American workers. It’s about as Pareto-optimal as legislation gets. So let’s open the borders, and encourage much more immigration into the US!

Salmon links to this report from the Federal Reserve Bank of San Francisco. Giovanni Peri, the author of the report, compared states with high levels of immigration to states with low levels of immigration, a sort of “natural experiment.”

For example, in California, one worker in three was foreign born in 2008, while in West Virginia the comparable proportion was only one in 100. By exploiting variations in the inflows of immigrants across states at 10-year intervals from 1960 to 2000, and annually from 1994 to 2008, we are able to estimate the short-run (one to two years), medium-run (four years), and long-run (seven to ten years) impact of immigrants on output, income, and employment.

Peri found that immigration is strongly beneficial:

First, there is no evidence that immigrants crowd out U.S.-born workers in either the short or long run. Data on U.S.-born worker employment imply small effects, with estimates never statistically different from zero. The impact on hours per worker is similar. We observe insignificant effects in the short run and a small but significant positive effect in the long run. At the same time, immigration reduces somewhat the skill intensity of workers in the short and long run because immigrants have a slightly lower average education level than U.S.-born workers.

Second, the positive long-run effect on income per U.S.-born worker accrues over some time. In the short run, small insignificant effects are observed. Over the long run, however, a net inflow of immigrants equal to 1% of employment increases income per worker by 0.6% to 0.9%. This implies that total immigration to the United States from 1990 to 2007 was associated with a 6.6% to 9.9% increase in real income per worker. That equals an increase of about $5,100 in the yearly income of the average U.S. worker in constant 2005 dollars. Such a gain equals 20% to 25% of the total real increase in average yearly income per worker registered in the United States between 1990 and 2007.

Basically, the US employment market is not a zero sum game. When immigrants come to the US to work, that benefits them (which is a strong reason, in and of itself, to favor opening the borders), but it also benefits us.

It’s also worth remembering that “life without competition with low-skilled non-Americans” is not an option on the menu. American workers will experience the downside of competition even if every single undocumented immigrant was somehow magically deported. But if those undocumented immigrants aren’t in the US, then Americans receive far less of the benefits. As the Economist blog wrote a couple of years ago:

Another possibility is that immigration also increases labour demand. This becomes especially important when we remember two other things. First, the one point upon which everyone can agree in this debate is that immigration substantially increases the productivity and earnings of the immigrants themselves. Secondly, we need to ask how the importation of low-skilled workers is different from the importation of goods produced by low-skilled workers abroad. Absent immigration, Mr Borjas would argue, wages would be higher in America and lower in trading nations. As such, price competition for tradeable goods would press down native worker wages.

Why is that important? Well, for one thing, it suggests that it’s difficult to separate cross-border flows of workers from goods. For another, when comparing outcomes, we need to remember that immigrants are still around whether they immigrate or not. In other words, immigrants might lower the wages of domestic workers, but immigrant consumption demand is much higher when they work on the American side of the border. If they do not immigrate, they’ll still be competing with native workers via imports of cheap products, but they’ll also be buying far fewer American goods, because they’ll be a lot poorer. It’s still difficult to know how things play out in the end, but one shouldn’t pretend that the proper comparison is a domestic labour market with immigrants versus one without.

Notably, even George Borjas — the best-known (and almost the only) economist arguing against immigration — calculates that in the long run, immigration has no effect on US worker’s wages overall (he predicts that immigration lowers the wages of high-school dropouts by nearly 5%, but raises the wages of other Americans, including those with only a high school degree). And Borjas reached his results not through empirical examination of what’s actually happened in the real world, but through an abstract calculation in which he considered only the downside of immigration, but didn’t account for the economic benefits at all.

So the worst-case scenario from Borjas is 1) Not all that bad (some workers gain, others lose, but overall there’s no difference), and 2) based on the dubious assumption that immigration provides no economic benefits to native workers. On the best-case scenario, we have strong benefits for everyone, immigrants and natives alike. And let’s not forget, fighting “illegal immigration” is expensive.

For that reason, it just makes sense to favor open borders (for all immigrants, I’d argue, except violent criminals) and immediate amnesty for all undocumented immigrants.

Further reading: The immigration category at the Ambrosini Critique. And the immigration category at Cardiff de Alejo Garcia. And the study from the Federal Reserve of SF, which is really pretty readable.