We exploit the change to the minimum school-leaving age in the United Kingdom from 14 to 15 using a regression discontinuity design to evaluate the causal effect of one more year of education on cognitive abilities at older ages. We find a large and significant effect of this reform on males'...