Yassine Badra (PHEAAC, Université des Antilles, FR), Bertrand Chopard (LIRAES, Université Paris Cité, FR), and Huizhong Liu (Wenlan School of Business, Zhongnan University of Economics and Law, Wuhan, China). “R&D cooperation and spillovers: When do research joint ventures improve welfare?”, Mathematical Social Sciences, Volume 139, 2026, 102489, https://doi.org/10.1016/j.mathsocsci.2025.102489.

 

 Lien vers la publication

 

Abstract

This paper examines the strategic implications of research joint ventures (RJVs) in an oligopolistic setting with R&D spillovers, focusing on their impact on social welfare. Using the well-known R&D model of d’Aspremont–Jacquemin (AJ), we analyze different R&D cooperation structures: the non-cooperative regime, the R&D cartel regime, the cartelized RJV, and the non-cooperative RJV. Our findings challenge conventional views on the efficiency of RJVs. While the model of Kamien–Muller–Zang (KMZ) consistently predicts that RJVs lead to lower social welfare, we show that in the AJ framework, an RJV can outperform both the standard non-cooperative and cartel regimes when spillovers remain below a certain threshold. This result underscores the importance of model-specific assumptions in assessing R&D cooperation policies and offers valuable insights for industrial policy and antitrust regulation in innovation-driven markets.

Keywords

R&D spillovers; R&D cooperation; Performance of RJVs