This paper investigates the impact of exposure to teacher stereotypes on student achievement in primary schools. We measure both implicit and explicit stereotypes using item-based questions available in the European Value Survey and the Gender-Science Implicit Association Test. By exploiting the random assignment of students to teachers with different levels of stereotypes, we show that the gender gap in math performance, defined as the difference between boys' and girls' scores on standardized tests, significantly increases in classes with math teachers with stronger gender stereotypes. Additional evidence suggests that this result is driven by girls who benefit from having teachers with a girls-math attitude.

In line with previous research, we do not find any effect of teacher stereotypes on student outcomes in reading.