This paper estimates the effect of human capital on countries' youth unemployment ratio by using country average scores from the Program for International Student Assessment (PISA) and the unemployment ratio of the same birth cohort. The identification strategy is based on variations in skills within country x year and across PISA cohorts. We estimate that a 1 standard deviation increase in reading literacy decreases the unemployment ratio by 1.2 percentage points, which is accounted for by a significant increase in employment. We find some evidence that educational and labor market institutions play a role in the association between skills and unemployment.