Fecha: febrero 2023
Mariano Mastrogiorgio & Martina Pasquini
ABSTRACT
Despite the growing interest in social innovation, the literature on this topic is limited both at the theoretical and empirical level. Accordingly, a controversial issue concerns the lack of a shared definition of social innovation and, consequently, the existence of a validated method to classify and measure it. In this paper, we focus on one type of intellectual property rights (IPRs), namely, patents, which well-reflect firms’ intangible assets. We propose an exploratory approach based on natural language processing to identify patents with a social-oriented content that can be assimilated to social innovation. We employ a random sample of about 3800 U.S. patents of Spanish inventors to test our methodology and provide novel evidence. Our main findings show that first, social innovation in patents is multi-faced but mostly concentrated in green technologies; and second, that patents with a high level of social content are the ones with a significantly higher number of forward citations—that is, more radical and likely with more value.