Spinoffs in context: entry and performance across different industries

Four major stylized facts about spinoffs have been identified by the literature: (i) spinoffs perform better than de novo entrants, (ii) there is an inverted U-shaped relationship between the age of a firm and spinoff formation, (iii) better parents generate more spinoffs, and (iv) better parents originate better spinoffs. These stylized facts hold in some industries (e.g., automobiles) but not in others (e.g., lasers, disk drives, and asbestos). Existing theories of spinoff formation and performance explain these stylized facts but not the differences across industries. Inspired by the history-friendly models of industry evolution, the article presents an agent-based simulation model in which technological and demand conditions contribute to determine both the emergence and the performance of spinoffs. We assume that three main factors characterize the emergence of spinoffs: first, spinoffs emerge out of innovation activities within the parents; second, spinoffs share knowledge with their parents; third, spinoffs have some degree of product differentiation with respect to their parents. The model is able to generate the stylized facts identified by empirical research in the automobile industry and also to replicate the regularities and exceptions holding in lasers, disk drives, and asbestos.