Skilled migration does not benefit sending countries...

The Nazis’ dismissal of Jewish scientists had a much greater negative impact on research output than did the wholesale destruction of buildings and equipment by allied bombers...

Written by Noah Carl.

In a recent viral tweet, Samo Burja wrote: “Few economists bite the bullet that if immigration is good for the countries gaining people, emigration is bad for the countries losing people.”

In response, the economist Noah Smith penned an article titled ‘Why skilled immigration (usually) benefits both countries’. Although “there might be a few cases in which we should actually be worried,” he notes, “there are both theoretical and empirical reasons to think that skilled emigration can help the economies of developing countries”. In other words, high-skilled emigration is not merely not harmful; it’s actively beneficial.

Smith identifies various mechanisms by which high-skilled emigration might raise the living standards of those left behind in the origin country: remittances; gains from trade; technology transfers; incentives to invest in new skills; brain circulation (migrants coming home after learning new skills).

To his credit, he does also identify mechanisms by which high-skilled emigration might lower the living standards of those left behind. For example, he notes that loss of human capital might mean “less entrepreneurship” or “degradation in the quality of government”. You don’t say.

Yet according to Smith, “the literature pretty strongly supports the “brain gain” idea” (the idea that high-skilled emigration is a net positive for human capital in the origin country). On Smith’s reading, “just one single benefit — brain gain from increased education — is often enough to outweigh brain drain in its entirety”.

How can this be the case? As Smith points out, several studies have found a positive association between high-skilled emigration and skill formation, such that countries with high rates of emigration often end up with more high-skilled people than they had before. The effect appears to work mainly through incentives to invest in new skills. When people know they might be able to move to a rich country, they’re more likely to acquire the necessary qualifications. Yet not everyone who wants to move is actually able to, so a lot of newly qualified people are left behind in the home country.

For example, Smith cites a paper which studied the emigration of Filipino nurses and found that “for each nurse migrant, 10 additional nurses were licensed”. He cites another paper which studied the emigration of Indian IT workers and found that “as the number of US visas were capped, many remained in India, enabling the growth of an Indian IT sector”.

I’m afraid I’m not convinced. Having more people with certain qualifications around doesn’t mean very much if those left behind are less intelligent and hard-working than the ones who left.

Economists have traditionally viewed “human capital” as analogous to physical capital like factories and computers; indeed, that’s where the very term comes from. Individuals “invest” in new skills, thereby becoming more “productive”. And while there is some truth in this view (learning to code really will make you a better IT worker) it is largely mistaken. Why? Because it ignores the role of genes.

We don’t live in a world where everyone starts out with the same baseline level of human capital, and then some individuals invest in more, while others invest in less. Rather, we live in a world where people start out with dramatically different levels of human capital – purely because of their genes. And for many skills, there is no amount of investment that someone born with low human capital could make that would enable them to acquire those skills. A sizeable percentage of the population cannot learn to code; an even larger share will never be nuclear engineers, and so on.

When we’re talking about individual differences in human capital, intelligence is the most important trait. And of the variation that’s not measurement error, developmental noise or random events like getting hit on the head, the vast majority is genetic. The proportion may be somewhat lower in developing countries – though less so among the cognitive elites of those countries, who have access to most of the same advantages as Westerners.

What this means is that treating “Indian IT workers” or “people with a university degree” as groups that reflect nothing more than skill investments is erroneous. The average member of those groups is much smarter than the average member of the population. So losing a large number of them means losing a large number of your most talented citizens.

Even within those groups, some individuals are much smarter than others. This is pretty obvious when you think about it. Which would be worse for the US? Losing 100,000 STEM graduates to China or losing 100,000 gender studies graduates? Clearly the former (the latter would probably be a positive).

Generally speaking, the marginal members of skill groups are the least smart.¹ For example, the average intelligence of US college graduates has been trending downward over time because less and less intelligent people have been enrolling in college. When only 10% of people go to college, they tend to be selected from the right-tail of the bell curve. When 50% of people do, they can’t be selected from the right-tail even in principle.

An economist sees a country lose 100,000 college graduates to emigration before gaining 200,000 through increased education, and concludes that human capital has doubled. But if those 100,000 were among the smartest and hardest-working people in the country, human capital has not doubled; it has almost certainly fallen.

Yes, increased education will raise the skills of those left behind (assuming education is less than 100% signalling). But this isn’t going to compensate for the emigration of people highly selected for intelligence. The loss of such people is harmful in two ways: it reduces the number of smart people around in the current generation; and it slightly lowers the population’s genotypic IQ, thereby reducing the number of smart people around in future generations.

“Okay,” Smith might say, “even if the net effect of emigration on human capital is negative, it’s unlikely to be very large. And it’s surely outweighed by things like remittances, gains from trade and technology transfers.”

Again, I’m not convinced. Intelligent, hard-working individuals are the single most important resource for a country’s economic development. And developing countries have fewer of them to begin with.

Some countries score more than two standard deviations below others on the World Bank’s measure of harmonized test scores – equivalent to an IQ gap of more than 30 points. (I’m assuming Smith doesn’t like national IQ data, which show the exact same thing.) The low-scoring countries therefore have many fewer people in the right-tail of the bell curve. And it’s precisely these people who are needed to start companies, to design infrastructure, to administer government – to do all the things that are necessary for economic development.

If you ask an economist why some countries are rich and some are poor, they’re likely to say “institutions”. But institutions don’t come from nowhere. They’re built by people – specifically, by people with high human capital.

While things like remittances, gains from trade and technology transfers are certainly beneficial, there’s no evidence that they’re important predictors of economic development. By contrast, human capital (measured properly) is consistently among the most important predictors.

A 2016 paper by the economist Fabian Waldinger is particularly illustrative. He obtained department-level data from German universities between 1930 and 1980, and found that the Nazis’ dismissal of Jewish scientists had a much greater negative impact on research output than did the wholesale destruction of buildings and equipment by allied bombers!

If you believe that “scientist” is a category that reflects nothing more than skill investments, this result is rather surprising. Couldn’t the dismissed scientists just be replaced by newly qualified ones? No, they couldn’t. The ones who were dismissed had exceptional intelligence and creativity, so losing them was a major hit to human capital.

Strangely, at the end of his article, Smith states that “we should take more of our skilled immigrants from large poor countries like India, Indonesia, and Nigeria” – presumably because certain benefits from emigration (like technology transfers) are largely independent of the number of emigrants, and because losing a given number of people makes less of a difference to a large country.

But if Smith is right that emigration “usually” benefits both countries, why should the US only take skilled immigrants from large poor countries. Why should it deprive small poor countries of the supposed benefits of high-skilled emigration?

The truth is that high-skilled emigration does not benefit poor countries. Rather, it harms them – by depriving them of the human capital they desperately need to develop. Unfortunately, not everyone can win: and in the case of high-skilled emigration from poor countries, it’s the people left behind who lose.

Noah Carl is an Editor at Aporia Magazine.

---

Consider supporting us with a $6.99 monthly subscription and following us on Twitter.

1. Smith seems to acknowledge this in his comments on the study dealing with Filipino nurses, observing, “it’s not clear that the new nurses are as high quality as the ones who left”.