Elements, characteristics, and performances of inter-enterprise knowledge recombination: Empirical research on green innovation adoption in China's heavily polluting industry

Knowledge recombination has become a hot topic in green innovation adoption (GIA) research in recent years. However, previous studies have mainly discussed the knowledge recombination within a single enterprise but ignored inter-enterprise knowledge recombination (Inter-KR). In this paper, a theoretical framework of Inter-KR is developed, and the three features of Inter-KR are discussed: elements, characteristics, and performances. Multisource heterogeneous data combined with Natural language process, network analysis and econometric regression are used to analyze the above three features. Social network theory is introduced to elaborate on the influence mechanism of Inter-KR. The empirical results indicate that enterprises can effectively improve their environmental performance by participating in Inter-KR: 1) the participation of Inter-KR is significantly influenced by time and national policy; 2) further refinement and decomposition of the characteristics of inter-KR find that the comprehensiveness has a significant promoting effect on GIA. When issuing relevant policies, the government should consider the balanced development of environmental governance and enterprise innovation, and improve the GIA performance by enabling enterprises to acquire an in-depth and extensive knowledge base. In summary, this paper provides new insights for research on knowledge recombination, the literature on collaboration innovation, and green innovation adoption research. [Display omitted] •Theoretic framework of inter-enterprise knowledge recombination was investigated.•Features of elements, characteristics, and performances were analyzed.•Natural language process, social network, and econometric model were applied.•Collaborative enterprises are more willing to adopt green innovation.•Inter-enterprise knowledge recombination promotes enterprise's performance.