Dynamic balancing of exploration and exploitation: The contingent benefits of ambidexterity

Abstract

We study the evolution of firms' exploration-exploitation allocations and their long-term performance outcomes. Extending current ambidexterity theory, we suggest that not only firms pursuing one-sided exploration or exploitation orientations show self-reinforcing tendencies but also ambidextrous firms adopting balanced exploration- exploitation orientations. Integrating formal modeling arguments,we further propose that reinforcing ambidexterity can be good or bad for firms' long-term performance, depending on the environment they face: In contexts characterized by incremental change, firms benefit more from the learning effects of maintaining ambidexterity, which lead to superior performance. Firms in discontinuous change contexts, however, suffer more from the misalignment that reinforcement creates, which affects their performance negatively. A longitudinal data set of global insurance firms (1999-2014) supports our arguments. Building on these findings, we reconceptualize ambidexterity as the ability to dynamically balance exploration and exploitation, which emerges from combining capabilitybuilding processes (to balance exploration and exploitation) with capability-shifting processes (to adapt the exploration-exploitation balance).We contribute to the organizational literature by developing a dynamic perspective on balancing exploration and exploitation, by clarifying the contingent nature of the ambidexterity-firm performance relationship, and by integrating and extending the ambidexterity and formal modeling perspectives on exploration and exploitation.