The paper adds to the scattered empirical evidence on the role of obstacles to innovation in a three-fold way. First, we correct for the usual sample selection bias by filtering out firms not interested in innovation from ‘potential innovators'. Second, we assess what mostly affects firms'...