Idiocracy: a review

Last night I watched Mike Judge's 2006 film Idiocracy. I enjoyed it a lot. It is basically a dystopian science fiction film with an element of (slightly dark) comedy, but it hides its fairly extreme pessimism and conservatism behind a half-hearted satire on modern society. It imagines what would happen if the less sharp people in society had substantially more kids than the smarter people—and if this trend carried on for centuries.

By 2505 the population has an average IQ of something like 60, and society is crude, degenerate and decadent. It limps on only because the last of technological advances went into automating most of the functions needed for basic survival. The personal narrative is that two people, a man and a woman both selected for their extreme averageness in all attributes, get cryonically frozen, wake up in 2505, and are the smartest people in the world. Hijinks ensue.

It's a fairly enjoyable film on its own merits, but the really interesting question is whether it says anything about the world we live in. While this sort of thing is extremely uncertain and speculative, arguably it does.

Behavioural genetics tells us, as the film suggests, that (variation in) intelligence is 50-90% driven by genes. And the extent to which intelligence is linked to genes increases through life (suggesting the impact of school & upbringing wears off quickly). This is true for any given socioeconomic class (unless they are exposed to lots of lead pollution, malnutrition or similar big negative environmental shocks). This is true across the world.

Indeed, a large fraction of this seems to come identifiably through specific alleles or single-nucleotide-polymorphisms (chunks of genetic code). We also know that intelligence correlates with brain volume, also genetically-driven. We even know that the way brain volume changes is substantially genetic. By contrast shared environment (includes family upbringing, and under certain conditions also the effect of school) has around zero effect on IQ/intelligence.

And so the question becomes: are sharper people having fewer kids than the less sharp? Evidence seems to say "yes", for the USA, UKTaiwan, and generally across the world. Though, it certiainly has to be said, that on the models here (estimating a loss of about .8 IQ points per generation, or perhaps 3 per century), it would take substantially longer than 500 years to get to the idiocratic society.

But wait a second—what about the Flynn Effect? Haven't measured IQs increased massively over the past hundred years? Aren't we getting smarter? Sadly, the Flynn Effect may not reflect an improvement in intelligence—once you account for the many ways that people have got better at tests, e.g. through learning to guess when they don't know the answer. In fact, the tests with a lower "g loading", i.e. the ones less good at accurately reflecting intelligence, are the ones reporting the biggest Flynn Effects.

This shouldn't be surprising, because random selection into a better school typically has no positive impact on achievement (though it does have strong negative impacts on crime). But even if the Flynn Effect were, say, reflecting greater education making up for lower genetic intelligence, it seems like we've pretty much exhausted its benefits—and it is now going into reverse in the Netherlands, theUK, Finland, and elsewhere. And thus even if phenotypic IQ (i.e. as measured by tests), and not genetic IQ, is "what matters", we can't hope that extra education will make up for duller genetics in the future.

And we have independent reasons to think that genetic IQ might be "what matters". While it is phenotypic IQ that correlates with, e.g. homicides, this relationship is seriously confounded by the fact that medical technology has advanced substantially over the period. Without these advances, homicides would be five times higher (on a static analysis—it's possible that the extra who would have been killed, but survived, would have themselves committed extra homicides).  And it is genetic IQ that appears to be associated with social advances, innovation, science, technology and so on. Were the Victorians smarter than us? Though the linked paper has been criticised, the authors' response to their critics is persuasive enough that we should take the idea seriously.

If this is all true (and certainly that is itself a contested step) should we be worried? This is the most interesting question for me. Thankfully, there are two families of technologies that I see as potentially solving this problem.

The first is artificial wombs. One of the main causes of reduced fertility among the smart, as documented in Idiocracy's excellent opening third, is the cost of having children. This cost is not just in terms of feeding, clothing and looking after them, nor even buying more expensive houses to get into better schools, but also in terms of labour market potential—much of the gender wage gap appears when women take time out to have & raise kids, and a large fraction of the rest may come because women expect to take this time out (and thus invest less in human capital). We know that people really do respond to things that make having kids cheaper. Technologies that drastically reducethe cost to smart women of bearing kids could be one way of arresting the alarming trend.

The second is genetic engineering in all its forms. If we can pick embryos, or even engineer people's DNA, then we could feasibly (after lots more research!) make sure that kids are smart even if their parents are not. While Idiocracy does not mention "my" first solution, it dismisses this one, explaining that the remaining smart people in society spend their precious time perfecting cures for hair loss and impotence.

But I'm much more optimistic than Mike Judge (hopefully rationally!)--I have almost unlimited faith in the ability of human ingenuity to overcome big threats to society. I feel confident we'll either develop the aforementioned technologies, or in their stead something completely different and entirely unpredictable that fixes the issue. And if our best minds do spend their time solving erectile dysfunction—as in Idiocracy—who knows, that might help solve the problem!

UPDATE 30/7/14: Jaymans says this trend may have reversed for the most recent cohort!

UPDATE 28/10/14: A paper newly published in the Economic Journal suggests that the education-fertility curve may have recently become U-shaped for US women.