Unveiling knowledge flows in the automotive innovation environment using patents and citations

Purpose \u2013 In the last years, researchers have shown that innovation processes are veering towards an \u201copen innovation\u201d approach whereby innovating firms rely increasingly on contributions by external partners (Chesbrough, 2003). In particular, the automobile manufacturing industry is a typically complex technical environment where the cooperation in knowledge creation is very extensive (Lin and Cao, 2012). On these basis, this paper aims to analyze if and to what extent, innovation output in terms of patent quantity and value differs depending on whether an automaker engages in a (certain type of) collaboration for its R&D activities. Design/methodology/approach \u2013 To reach our aims, we selected a quali-quantitative approach based on the collection of two datasets about four among the world largest automakers: the former contains a total of around 4,400 relationship pairs of automakers and suppliers, allied companies and subsidiaries; the latter includes the automakers\u2019 EPO patent portfolios, and related information, since 2004 which provide the names of the co-assignees. Matching the company names of the two data sources allowed us to obtain the final dataset used to carry out the extensive descriptive analysis and regression analysis. Originality/value \u2013 The paper contributes to the scientific literature by investigating the presence of differences across the patenting activities of a selection of automakers and their supplier, allied and subsidiary firms. Moreover, the issue related to patent value represents an emerging area of interest in the field of collaborations for innovation. Also the methodology constitutes a novelty in this field, since the examined database is generated by matching two different sources. Furthermore, the process for cleaning and standardizing the company names (\u201cname game\u201d) combined an automated algorithm with a double manual check, by searching the company websites and querying the corporate tree. Practical implications \u2013 The outcomes of the study show that the four selected automakers adopt different strategies regarding co-patenting with their R&D partners. On this basis, a framework including four main strategies -contingent, purposive, watchful and advanced- have been identified and lays the foundation for further research in this field. In addition, the present study reveals that co-patented inventions with suppliers and subsidiaries seem to be more valuable than with allies. Evidences emerging from our paper may support firms in making aware choice regarding the management of knowledge collaborative activities and co-patenting