Garett Jones — Immigration, national IQ, & less democracy

Dwarkesh Patel asks whether The Culture Transplant is really a continuation of Garett Jones’s case against democracy. Jones answers that citizens shape governments and productivity in almost any regime; even in China the government “has to listen to the masses.”

Asked to split the effect of migrants’ votes from the effect of their presence, Jones offers an estimate rather than a measurement: roughly 50-50. Citizens also staff the firms that build or fail to build the complex exchange networks behind high productivity.

I’ll cheat a little bit because we don’t get to run experiments on this, so I just have to make an informed guess.

Jones thinks the upper tail of ability matters more than the lower tail within the normal range. His reasons: moderately prosperous nations have tools to reduce the influence of the least informed voters, and crime from the low end is a probabilistic risk, not an on-off switch.

He finds the median-versus-elite contrast more interesting. Democracies should be driven by the median, yet the exceptions where mean scores predict productivity best are non-democracies and South Africa, plus Gulf states such as Qatar and the UAE with highly educated migrant communities.

Patel asks whether a wider IQ spread at a fixed mean would help. Jones says yes in the normal range, because elite breakthroughs create positive externalities, a view he calls almost Pollyanna-ish Paul Romer growth theory. Applied to low-skilled immigration, he rejects the premise.

You didn’t just increase the variance, you simultaneously increased the variance and lowered the mean and median, right?