In several countries, students who fail end-of-high-school high-stakes exams are faced with the choice of retaking them or forgoing postsecondary education. We explore exogenous variation generated by a 2006 policy that imposed a performance threshold for admission into postsecondary education in Greece to estimate the effect of retaking exams on a range of outcomes. Using a fuzzy regression discontinuity design and novel administrative data, we find that low-achieving students who retake national exams improve their performance by half a standard deviation, but do not receive offers from higher quality postsecondary placements. The driving mechanism for these results stems from increased competition.