Finance Seminar - Gordon Phillips

Title: Collaborate or Consolidate? A NLP Text-Mining Analysis of R&D Networks and M&A
Authors:  Chih-Sheng Hsieh, Michael D. König, Gordon M. Phillips, and Jakob Rauch
Abstract:  We build a novel dataset of R&D collaborations between firms by text mining news articles and using the RoBERTa transformer model to classify collaborations. Our data set covers several decades and represents the largest currently existing database on R&D collaborations. The comprehensiveness of our data allows us to document differences in R&D collaboration activities across firm characteristics, industries, countries, or regions and how these differences change over time. We then estimate a structural model capturing the joint endogeneity of both R&D collaborations and M&A activities to identify technology spillovers and market competition effects in firm performance and how these change over time with the evolution of technologies and markets. Our results indicate that the changing pattern in technology spillovers is due to firms substituting R&D collaborations with M&A, emphasizing the close connection between R&D collaborations and M&A activities.