Using Machine Learning to Uncover the Role of Outside-Industry Catalysts in Multi-party Alliances

Abstract

Participating in a multi-party alliance with competitors is an important yet tricky decision for firms. By focusing on interactions with industry peers, the extant literature has largely ignored the role of outside industry participants. We posit that industry outsiders impact this decision by behaving as catalysts that influence the governing principles and the milieu that underlies the alliance. While some catalysts induce an environment of formal governance that reduces inter-partner trust and disincentivizes collaboration among industry rivals, others create an opportune environment for learning that supports collaboration. Given little theoretical guidance by prior literature, we employ a novel methodology that uses pattern recognition by machine learning algorithms to inductively develop and test our theory by analyzing heterogenous deal syndicates in the private equity (PE) industry.