Contact theory and conflict theory offer sharply conflicting predictions about the effects of interethnic exposure on prejudice. Contact theory predicts that close collaborative contact under conditions of equal status causes a reduction in inter-ethnic prejudice. By contrast, conflict theory predicts that shallow or competitive exposure causes an increase in inter-ethnic antipathy. Both theories are backed by rigorous field-experimental evidence. However, much of this evidence tests each theory under arguably extreme conditions. Therefore, the boundaries of the scope conditions for contact and conflict theory remain unclear: where is the line between close versus shallow, or collaborative versus competitive, inter-ethnic exposures? We test the consequences of inter-ethnic exposure in a natural and non-extreme setting by executing a large, well-powered, and pre-registered randomized field experiment on inter-ethnic discrimination in 40 Hungarian schools. We show that neither manipulating the closeness of interethnic contact within classrooms, nor variation in inter-ethnic exposure across classrooms, has an effect on non- Roma students’ inter-ethnic discrimination. These findings suggest that inter-ethnic contact may be neutral with respect to discrimination in many everyday settings, thus failing both to fulfill the hopes of contact theory and to actualize the concerns of conflict theory.