I think the bigger question, since it apparently involves reported bullying incidents rather than teacher-observed ones, is whether the change in policy results

I think the bigger question, since it apparently involves reported bullying incidents rather than teacher-observed ones, is whether the change in policy results in underreporting or overreporting of bullying. Plausibly it could be a lot more than a 13% fall if kids feel much more incentivised to report it, or a negligible effect if there's a 13% chance of it not being reported because bullied kids don't want to get their bystander classmates into trouble...