We identify earnings impacts of exposure to an infant health intervention in Sweden, using individual linked administrative data to trace potential mechanisms. Leveraging quasirandom variation in eligibility, we estimate that exposure was associated with higher test scores in primary school for boys and girls. However only girls were more likely to score in the top quintile. Subsequent gains, in secondary schooling, employment, and earnings, are restricted to girls. We show that the differential gains for women accrued from both skills and opportunities.