Private and Social Learning with Endogenous Timing: An Experimental Analysis

Abstract

We conducted laboratory experiments to investigate how private and public information affect the selection and timing of technology adoption. Our experiments extend the standard herding model to more accurately represent the innovation decision problem. Subjects drew private signals and observed actions of their group before making an irreversible choice between a safe and a risky innovation. Free to choose the timing of their adoption, equilibrium behavior dictates adoption of the innovation favored by the first private signal. Nevertheless, roughly half of subjects delayed adoption beyond the first round. When they did adopt, subjects gave more weight to their private signals than to their peers' actions. The speed and accuracy of adoption decisions improved when subjects observed their peers' decisions, even when subjects' payoffs were statistically independent --- as if observation exerts peer pressure'' on subjects. Finally we examined several plausible behavior rules and conclude that subjects find it profitable, on average, to wait until the second round and then follow their private signals.