'Global Inequality as a Consequence of Human Diversity'

EDIT: A version of this review was published in the journal Intelligence and can be found here. Over the past few days I read Tatu Vanhanen's new book, Global Inequality as a Consequence of Human Diversity published by the Ulster Institute for Social Research. Though I am broadly open to the arguments he makes therein, it is not a good book and I cannot recommend it. I have not read IQ and the Wealth of Nations the 2002 book he published with Richard Lynn, but I suspect the theses are very similar.

His argument runs that the existing explanations for the variation in living conditions between nations are either hard to falsify (like Jared Diamond's and Jeffrey Sachs'), excessively narrative, explain too little of the variation, or beg the question (like Daron Acemoglu & James Robinson's).

He advances an alternative explanation: since (a) human evolution has been recent and extensive enough to produce substantial morphological differences (consider how different Koreans, Somalians and Papua New Guineans look); and (b) we have a large literature suggesting that lots of traits, especially cognitive ability (measured well, he believes, by IQ) are substantially genetic, we shouldn't assume all human populations are genetically similar in intelligence. Since we know that intelligence is important for social success within populations we should not be surprised if it controls success across populations.

Quoting Pilar Ossorio, he says:

According to contemporary geneticists, any two unrelated humans are about 99.8% or 99.9% genetically identical, but because 'the human genome contains approximately 3bn nucleotides (DNA building blocks), a 0.1% or 0.2% difference translates into millions of sites at which two people will have a different nucleotide.'

He constructs an 'Index of Global Inequality' (IGI) out of six measures of societal success: gross national income at purchasing power parity; the share of the population in tertiary education; under-five mortality; life expectancy; sanitation facilities; and an index of democracy. He finds, with a simple regression, that around three quarters of a country's IGI is explained by its average IQ. Further, most of the outliers have special explanations for why they depart from the curve, the biggest four being: oil, caribbean tourism, history of socialism and civil wars.

In and of itself, this argument is somewhat interesting and somewhat suggestive. What is surprising is that this is almost the entirety of the book, fairly laboriously spread over 170-odd pages. It makes almost no attempt to deal with the evidence against it. For example, it's true that within Western groups IQ is highly heritable (i.e. most of the variation between individual is due to genetic factors) and largely unaffected by environment. It's also true that within poorer societies a large fraction of IQ is down to genes, but this fraction is lower, because there are many more of the big downside factors that can really stunt cognitive development (principally malnutrition). And it's much less clear that the difference in IQ across the world, where environments are extremely heterogeneous, are down to genetics.

It's not that it's impossible they are and there is rock solid evidence for the alternative. But Vanhanen doesn't even attempt to provide evidence for his view that IQ is practically entirely genetic, even across strikingly different environments. One flaw of the book is that once he has stated that the evidence suggests IQ is mainly genetic within populations, he begins assuming the link between IQ and development across populations purely represents human diversity-caused differences, with only the unexplained residual environmental. This is not representative of the stronger stuff I've read in the area.

This leads into another puzzling issue with the book: the fact it fails to deal with any of this existing literature. There are better ways to test whether institutions, geography or human diversity is driving differences in development and he doesn't really attempt any of them! For example, a 2011 paper found that effectively random variation in institutions had no effect on the economic outcomes of a given African ethnic group. The literature is large and the debate is still raging, and Vanhanen's 'side' might end up being judged right, but his approach adds basically nothing to the question of whether IQ causes development or development causes IQ.

I suppose I can't really blame the book for its cheapness, the relatively frequent typos, the ugly and unclear charts and tables, but these certainly reinforce the overall feeling of lightweight pointlessness you get when you read it. And I can't stress enough the importance of human capital theorists being epistemically cautious in their claims, given how controversial their conclusions are. People, working on sound principles, often devalue a perspective when they hear a weak argument in its favour ('if that's the best they've got...'), and I think Vanhanen's new book will only weaken the case for considering human diversity when looking at global inequality.