This study examines the impact of children's age within grade on school victimization in elementary school. Identification of age effects relies on the instrumental variables approach drawing on official school entry age rules based on children's month of birth. The empirical analysis uses the PIRLS data for 17 countries where such school entry age rules are effectively applied. Possible selection into compliance with official entry rules is taken into account via a control function approach. The study demonstrates that children are causally and significantly harmed by being the youngest within grade. Sub-group analysis reveals that the size of age effects on school victimization tends to be higher for boys than for girls as well as for children with an immigrant background compared to natives. The point estimates suggest that the age effect on school victimization is especially high in countries where there are also high effects on the cognitive outcome variable.