This is related, but not the research talked about. The Terman Project apparently found that the very highest IQ cohort had many more patents than the lower cohorts, but this did not show up as massively increased lifetime income.

Compare the bottom right IQ graph with SMPY results which show the impact of ability (SAT-M measured before age 13) on publication and patent rates. Ability in the SMPY graph varies between 99th and 99.99th percentile in quintiles Q1-Q5. The variation in IQ between the bottom and top deciles of the Terman study covers a similar range. The Terman super-smarties (i.e., +4 SD) only earned slightly more (say, 15-20% over a lifetime) than the ordinary smarties (i.e., +2.5 SD), but the probability of earning a patent (SMPY) went up by about 4x over the corresponding ability range.

http://infoproc.blogspot.com/2011/04/earnings-effects-of-personality.html

Unless we want to assume those 4x extra patents were extremely worthless, or that the less smart groups were generating positive externalities in some other mechanism, this would seem to imply that the smartest were not capturing anywhere near the value they were creating - and hence were generating significant positive externalities.

EDIT: Jones 2011 argues much the same thing - economic returns to IQ are so low because so much of it is being lost to positive externalities.