We study whether education raises productivity (human capital theory) or just reflects it (sorting theory) by estimating returns to a high school diploma. For this purpose, we exploit standardized exit exams students must take in the final year of secondary education to obtain a diploma in the Netherlands. Using a regression discontinuity design on administrative population data, we compare earnings of students who barely passed and barely failed standardized exit exams. The results indicate a positive effect of a high school diploma on earnings of about 0.34 EUR per hour. We interpret this finding as evidence of diploma sorting effects.