Abstract
‘Recombination’ and ‘impact’ have become well established constructs to understand the origins of inventions and their importance for the development of future inventions. Despite forming these two familiar ‘faces of inventions’, their specific relationship has only marginally been subject to inquiry. To address this, this paper studies the relationship between the level of recombination of inventions and their technological impact, along two steps. First, in contrast to the common idea of a linear relationship between recombination and impact we argue that the relationship is in fact a non-linear one. Second, we distinguish between different levels of recombination (low, intermediate, high) and determine their differential impact, thereby establishing which type of recombination leads to the highest level of technological impact. We test our hypotheses on an extensive dataset, comprised of all USPTO granted patents in the biopharmaceutical industry between 1976 and 2006. Our empirical findings indicate strong evidence for a curvilinear relationship between recombination and impact. In addition, we find that an intermediate level of recombination – formed by a combination of components from local, adjacent and distant knowledge domains – carries the highest level of technological impact of all types of inventions. Finally, we discuss implications for the academic literature and for firms’ innovation strategies.
