16385: Exploring the Risk of Biotech–Pharma Alliance Breakups Following Drug Launches (session 26286)

Biotech-pharma alliances are essential to drug development but often involve asymmetrical dependencies, with biotech firms more reliant on their pharma partners than vice-versa. This imbalance can pose an overlooked vulnerability to biotech firms: an increased likelihood of alliance termination following a pharmaceutical partner’s drug launch. Using a dataset of more than 1,600 terminated biotech-pharma alliances spanning 20 years, we find that the risk of alliance termination increases post-launch. Using machine learning models and explainable artificial intelligence (AI) methods, we show that this risk is predictable at the onset of an alliance and demonstrate the importance of considering complex interactions and nonlinear effects. Our findings suggest that these terminations are not simply unpromising alliances strategically delayed to offset the negative impact of a termination announcement with the positive news of a drug launch. Instead, they reflect an unexplored cost of open innovation in asymmetric alliances and provide insights for biotech firms to effectively evaluate prospective partners against the risk of an alliance termination after a drug launch. Methodologically, we show the value of machine learning and explainable AI in management research to explore previously unexamined phenomena through a data-driven approach.

view paper (if available)