Cities! Education!

"Our instrumental variables estimate is that on average, an extra year of schooling [city average] is associated with 22.0% higher hourly wages across cities." (Glaeser and Lu) Their methodology, if valid, implies that this is a causal effect of human capital spillovers, not just productive cities building more schools or attracting better-educated migrants.

A human capital spillover of this magnitude seems enormous to me, given that the typical estimate of the individual-level return to a year of schooling might be on the order of 10%. It's as if a city gets a 200% bonus on the return to schooling because smarter people get to interact with other smarter people.

The authors note that although huge, their estimate doesn't come close to explaining the bulk of China's growth in per capita income over the past three decades. Still, it's pretty good news for education- and city-boosters... like me!