Researchers in academia typically perform different tasks: research, teaching and services to the scientific community. We analyze the opportunity costs in terms of a potentially reduced publication productivity associated with becoming a dean in the German institutional setting where deans are non-professional expert-leaders who temporarily take the dean position. Theoretically, we distinguish between two different effects that relate deanship and publication productivity: a resource effect where publication productivity during and – as a result of potentially having developed a taste for service –also post deanship decrease as a result of a reduction of the available time for research and a self-selection effect where pre-deanship publication productivity is lower than that of peers who are not about to become dean. Based on a dataset of 1,110 business and economics researchers from German-speaking universities, we find evidence for a resource effect with leadership in academia reducing research productivity during and also post deanship. We find no evidence of a negative self-selection effect in the sense of less successful researchers being more likely to take the position of a dean. Reduced research productivity during and post deanship as compared to those researchers that never became dean is driven by those researchers who become dean in later periods of their career, i.e., presumably by those who deliberately shift their focus away from research and towards a stronger engagement in the scientific community in their late career years. Early career deans, on the contrary, seem to see their deanship more as a transitory role and are able to compensate the reduced resources during deanship, and they also do not suffer from a reduced publication productivity post deanship.