Previous literature on the effect of tenured and tenure-track vs. non-tenure-track professors on students' performance at university finds contrasting results. Our paper is the first to test whether tenured/tenure-track and non-tenure-track teachers differently affect students' performance at school. We use data on standardized test scores of a representative sample of primary and secondary school students in Italy and information on their Italian and mathematics teachers' labor contracts. Controlling for class- and subject-fixed effects, we find that non-tenure-track teachers decrease students' performance by 0.21 standard deviation. This detrimental effect is fully explained because non-tenure-track teachers are less experienced. In line with previous findings on the adverse effects of teachers' absences, non-tenure-track teachers are also associated with 0.1 standard deviation worse student performance when their contracts last less than a year.