Decomposing Uncertainty and Its Effects on Imitation in Firm Exit Decisions

This study examines the effects of different uncertainty types on interorganizational imitation in firm exit decisions. We draw on herding models to conceptualize exit decisions as being based on a firm’s private information, which the firm updates with information inferred from observing the actions of others. We posit that different types of uncertainty differentially affect this observational learning process; in particular, we propose that certain uncertainty types attenuate (rather than foster) observational learning and subsequent imitation. We test this theory using a 29-year panel data set on the exit of private venture capital firms. Our results indicate that observational learning does influence imitation in firm exit decisions, and they also suggest that a common belief—that uncertainty enhances imitation—does not apply to all types of uncertainty. Specifically, we find that uncertainty fosters imitation only when it is idiosyncratic to the firm; uncertainties that are common to all firms, in contrast, actually reduce reliance on observational learning. By decomposing uncertainty into different types and explicating their effects on imitation, we demonstrate that this relationship is more nuanced than previously assumed and, in addition, highlight the role of deliberate information processing in imitation.