Wednesday, July 04, 2007
The Consequences of Brain Drains in Developing Countries
"They say there is a brain drain. Let these decayed brains flee. Do not mourn them, let them pursue their own definitions of being. ... Don't be concerned. They should escape. [Iran] is not a place for them to live any more. These fleeing brains are of no use to us. Let them flee. If you know that this is no place for you, you should flee too."
- Ayatollah Ruhollah Khomeini [English, Farsi]
Recently, Steve wrote that he was surprised I had not addressed brain drains (the flight of the educated elite from less productive to more productive countries) before. So in this small post, I will. Specifically, I will address the Brain Drain among developing countries, using as examples Ethiopia, China, and Iran.
Intelligence is heritable. When living standards are relatively close, about 50% of variation in intelligence is explained by genetics, about 10% by home environment, and about 40% by "individual differences" (in statistical jargon, the error term). In the United States, where standard of living that much of home environment is itself driven by genetics, intelligence seems to be 70% heritable. In poor countries with large class differences, however, home environment counts for a lot. This matters because the more of intelligence that is explained by genetics in a country, the worse the long term consequences of the brain-drain will be.
Thus, the brain drain has minimal long-term consequences for Ethiopia. Ethiopia is a largely agricultural, tribal-based society where individual differences in intelligence doesn't determine much. You are much more likely to be well fed and have children with wives/mistresses if you are well connected than if you are smart. The intellegentia of Ethiopia is largely determined by political connections, not DNA.
Likewise, the brain drain has minimal long-term consequences for China. Among other reasons, China is a huge country that has a greater population than Europe and the Western Hemisphere combined. A statistically important fraction of China's population is simply unable to leave, as the loss of even .02% of China's population would be enough to double the population of Canada. And even apart from this, China has about 900,000,000 peasants who struggle to survive, let alone achieve the higher education that marks one as intelligent.
However, the brain drain has drastic long-term consequences for Iran. Iran's descent system of public education allows most Iranians a fair shake at greatness. Iran's objective college entrance tests root out corruption and nepotism in the admissions process. In all likelihood, 50% of the variation in intelligence is heritable in Iran, just as it is in other modern countries.
"Every year more than 150,000 educated young people leave Iran for countries such as the U.S. and Canada. Some 4 million Iranians now live abroad." At this rate, within a decade 10% of Iran's population will no longer live in Iran. Unless a massive flight back to Iran takes place, these expatriates and their families will become assimilated into western society and join the rest of the melting pot, losing a desire to return to the old country.
It would be an exaggeration to say that the top 10% of Iranians will have left... But perhaps not much of one to say that a third of the top 30% will have fled. While the drop in Iranian fertility will allow more of society's resources to be focused on fewer children, the genetic loss the brain drain causes will be around for generations. Low intelligence is behind much of Africa's problems of state-failure and institutionalized misery. The longer the brain drain continues in Iran, the dimmer Iran's future will be.
In conclusion: The Brain Drain presents long-term troubles as it reduces the genetically factors that lead to high intelligence. This is only a problem in states where living standards are comparable and good systems exist to educate the general population. Iran's successes in providing a decent life and good educational possibilities for its citizens make her future worse than those of her more corrupt fellow states.
"When living standards are relatively close, about 50% of variation in intelligence is explained by genetics, about 10% by home environment, and about 40% by "individual differences" (in statistical jargon, the error term)."
So, when comparing similar home environments, home environment doesn't change intelligence? I remain unconvinced of the premise of this post.
Posted by: a517dogg | Wednesday, July 04, 2007
"So, when comparing similar home environments, home environment doesn't change intelligence? I remain unconvinced of the premise of this post."
An insightful criticism. The .5 figure comes from nationwide studies in developed countries (that is, half of variation in intelligence among Americans is explainable by genetics, half of the variation in intelligence among Swedes is explainable by genetics, etc.).
Within a developed country, there aren't giant differences in nutrition, diseases, etc. Even when comparing intercontinentally, home environments are considered to be similar (American Japanese score similarly to native Japanese, American Germans score similarly to native Germans) except when nutrition or disease rates are very different (African-Americas score a full standard deviation above native Africans).
Raising the .5 figure even farther is "genetics expressed through the environment." For instance, children with books in the home read more, and children with intelligent parents read more. However, intelligent parents are more likely to have books in the home. If I recall correctly, adding in "genetics expressed through the environment" to the genetics quotient raises the explainable power of genetics to .7.
Posted by: Dan tdaxp | Wednesday, July 04, 2007
This is some fascinating insight. I am curious if you have any links handy for the studies you are referring to.
Posted by: TDL | Wednesday, July 04, 2007
The finding that intelligence is largely heritable is relatively new -- as of "A Candle in the Dark," Sagan argued against it and had consensus on his side.
Some older references (meaning 1990s) are available from madsci.org and the New York Times 
An overview of the controversy with some nice graphs is available from wilderdom 
More recent citations can be found from gnxp [4,5,6,7,8]
A longer treatment of the subject, with an exhausting list of references, is in Steven Pinker's "The Blank Slate" 
Some of the current links up the heritability estimate to .8, but my suspicion is that they are counting genes expressed through the environment, which is kind of flaky. Also note some findings indicating that genetic influences counts for more in higher class families, which makes sense, as these would be the least likely to be exposed to toxins, under-nutrition, etc.
Posted by: Dan tdaxp | Wednesday, July 04, 2007