I use differences in educational attainment by birth cohorts to estimate the rise
in the return to education in the United States. If average ability is similar among
nearby cohorts, then differences in educational attainment lead to differences
in earnings only if education is productive. The results reveal that (i) the return
to a year of schooling increased from 4.8 percent to 8.4 percent between 1964
and 2003, (ii) the ability bias rose from 1.8 percent to 4.7 percent during the
same period, and (iii) the acceleration in the education premium after 1980 is
explained almost entirely by the rise in the ability bias.