Don't forget the https://en.wikipedia.org/wiki/O-ring_theory_of_economic_development https://en.wikipedia.org/wiki/O-ring_theory_of_economic_deve... - Kremer, i

Don't forget the https://en.wikipedia.org/wiki/O-ring_theory_of_economic_development https://en.wikipedia.org/wiki/O-ring_theory_of_economic_deve... - Kremer, incidentally, won the Nobel this year (albeit not for the O-ring model). As Shockley and innumerable analyses have shown, 'ability' defined very narrowly may be normally distributed, but because tasks draw on many abilities and steps can multiply rather than add ('there is nothing so useless as doing well that which should not be done at all'), it's very easy for a bunch of normally-distributed abilities to yield a log-normal distribution of outputs, in which case there is massive inequality in final results: https://www.gwern.net/docs/iq/1957-shockley.pdf https://www.gwern.net/docs/iq/1957-shockley.pdf