We study whether raising instructional time can crowd out student pro-social behaviour. To this end, we exploit a large educational reform in Germany that has raised weekly instructional hours for high school students by 12.5% as a quasi-natural experiment. Using a difference-in-differences design, we find that this rise has a negative and sizeable effect on volunteering, both at the intensive and at the extensive margin. It also affects political interest. There is no similar crowding out of scholastic involvement, but no substitution either. Impacts seem to be driven by a reduction in available leisure time as opposed to a rise in intensity of instruction, and to be temporary only. Robustness checks, including placebo tests and triple differencing, confirm our results. [This paper was produced as part of the Centre for Economic Performance's Wellbeing Programme. The project was funded by the German Ministry for Education and Science.]