We conceptualise social entrepreneurship as a source of social capital which, when present in the environment, enhances commercial entrepreneurship. We also argue that social entrepreneurship should be recognised as a second form of Baumol's (1990) productive entrepreneurship and that it will therefore compete at the individual level for resources with commercial entrepreneurship. Unlike institutional void theory, we see social entrepreneurship as conditional on institutional quality, but consistent with the institutional void perspective we see it as filling the gaps where government activism is lower. These arguments motivate our hypotheses that we test and largely confirm applying multilevel modelling. Our analysis is based on population-representative samples in 47 countries (the 2009 GEM dataset).