Immigration externalities

Consider a country, with population average IQ of 100, that opens its borders and the population doubles as people with a average IQ of 80 enter from a neighbouring country. The populations of the United States and United Kingdom have average IQs around 100, while the populations of countries such as India and Central America have average IQs around 80. There is generally more than an order of magnitude difference in per capita GDP between the populations with 80 IQ and 100 IQ.

Following the opening of their borders, the average IQ of the population of the open borders country has dropped to 90. Country populations with estimated average IQs around that range include Greece, Turkey and Croatia. These countries have a GDP per capita of roughly 1/3 of the United States.

Will the GDP per capita of the open borders country move to a similar level of GDP per capita as other 90 IQ countries? Could we craft an argument that the existing institutions of the 100 IQ country would protect it? What would distinguish this country in the long run from other 90 IQ countries? If the per capita GDP will plunge, should immigration be restricted to high IQ people?

These sort of question often attracts eugenics insinuations, as in a post titled Economics, Immigration and Eugenics by Will Wilkinson. Wilkinson raises the eugenic views of many prominent (and “progressive”) economists in the early 20th century, and quotes Thomas Leonard, who wrote:

Irving Fisher [a very important figure in the development of 20th century economics] said in his presidential address to the Eugenics Research Association, “I should, as an economist, be inclined to the view that unrestricted immigration . . . is economically advantageous to the country as a whole . . . .” But, cautioned Fisher, “the core of the problem of immigration is . . . one of race and eugenics,” the problem of the Anglo-Saxon racial stock being overwhelmed by racially inferior “defectives, delinquents and dependents.”

Wilkinson then asks:

[W]hat would a latter-day Irving Fisher, with access to the best and latest science, say about immigration? I’d guess he’d be a big enthusiast for offering visas and citizenship to high-IQ, high-skill foreigners. But what about poorer, low-skill workers? I’m not so sure.

Wilkinson threw to Garett Jones for comment, who answered Wilkinson’s question in the following series of tweets:

High human capital immigrants are likely better voters, better savers, more Coasian.

& for incompletely understood reasons, human capital differences persist across generations & across immigrant groups..

…so an atheoretical forecaster would bet high IQ immigrants would probably yield a bigger long run social surplus…

…while lower IQ immigrants have to yield a Ricardian benefit > negative political externalities to raise surplus.

This argument reflects previous points made by Jones, such as in this paper in the Asian Development review (pdf). The paper provides a great summary of how IQ affects productivity and economic growth and deals nicely with some old chestnuts about the validity and usefulness of IQ testing.

Two factors that Jones did not directly mention in his tweet were that high IQ people tend to be more cooperative and trusting (and trustworthy), and that skill complementarities may be important in fragile, high-value production technologies.

Applying these points to the above thought experiment with the open borders country, what are the policy effects through voting of a lower IQ population (although this is possibly the easiest solved through allocation of voting rights)? How do social interactions and institutions change if there are more untrusting and untrustworthy people?

One factor in answering these questions might be the distribution of IQ in the new population. Assuming the population of each group before the immigration event was normally distributed, the new population would have more high IQ individuals (say, above 115 IQ) and more low IQ individuals (say, below 75 IQ) than you would expect in a typical population of average 90 IQ. This dispersion might decrease over time, but if mating is assortive, it could persist. If per capita income is highly dependent on the productivity of the best and brightest, the presence of a significant proportion of the population with high IQ could keep incomes high. If a small proportion of low-IQ individuals damages social trust, the immigration could have negative consequences beyond that expected in a typical country with average population IQ of 90.

I am not sure where the balance between Ricardian benefits and negative externalities from low-IQ immigration lies. The large changes in per capita income associated with changes in average population IQ suggests that this is not simply a case of economic growth edging down with lower productivity workers, but rather that there may be sharp changes in productivity where political and social institutions change, or where complex production processes become infeasible. Low income in developing countries is often explained as a result of poor institutions, but to what extent are those institutions endogenous to the country’s population? These potential externalities tend to be ignored in most stories of the Ricardian benefits of immigration.


  1. Professor Seymour Itzkoff has written about this in his book “The Decline of Intelligence in America”.

    Aside from intelligence, there may be other temperamental differences – for example in respect of impulsivity, conscientiousness and executive function. Even controlling for intelligence there are differences in things like group crime rates – which in turn has a negative impact on property values and schools as middle classes move away.

  2. Could we craft an argument that the existing institutions of the 100 IQ country would protect it?
    In the LR Yes, assuming that population IQ is highly mallable and that the better institutions (better education at the higher end , better childhood nutrition at the lower end ect.) in the high IQ could maintain the IQ at 100. Thereby eliminating any reduction in GDP per capita due to a expoected decrease in IQ.
    The difference in quality of education and access to an education of quality is where I assume it’d fall down.
    It would interesting to see if the IQs of the children of immigrants from lower IQ countries is closer to the total population mean when access (and quality) of education is controlled for.

    1. That’s a perfectly valid argument, but which parts of it will hold. Is IQ malleable? Can institutions affect IQ? If so, what are the pathways? How robust are institutions while the new arrivals transition? We can’t simply assume the answers to these questions.

  3. not just about IQ. We keep hearing that education is what is needed to lift people out of poverty and radicalism however, many of the 911 bombers were educated and probably had reasonable IQs. IQ does not matter if belief systems are dysfunctional.

Comments welcome

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s