Safety crises and R&D outsourcing alliances: Which governance mode minimizes negative spillovers?

This study examines how R&D alliance governance affects both the probability and magnitude of negative spillovers triggered by a partner’s safety crisis. I show that hierarchical governance leads to a lower probability that a partner will suffer a crisis and thus trigger a negative spillover, yet this governance mode leads to negative spillovers of greater magnitude should they happen. Because expected spillover costs are calculated as the probability of occurrence times the magnitude of such costs, it is not clear which governance mode best minimizes expected spillover costs. I combine transaction costs economics and signaling theory to develop a contingency model that identifies which effect is more likely to dominate, as a way to address the reported dilemma. I found evidence in support of the described dilemma and the contingency model in a sample of 296 R&D outsourcing alliances in the biopharmaceutical industry.