Low and highly volatile growth define Africa's growth experience. But there is no evidence that growth volatility is associated to long term economic performance. This result may be misleading if it suggests that volatility is not important for economic and social progress. In this paper we use a variant of the method developed by Hausmann, Pritchett, and Rodrik (2005) to identify both growth acceleration and deceleration episodes in Africa between 1975 and 2005. The authors find that Africa has had numerous growth acceleration episodes in the last 30 years, but also nearly a comparable number of growth collapses, offsetting most of the benefits of growth. Had Africa avoided its growth collapses, it would have grown 1.7 percent a year instead of 0.7 percent, and its GDP per capita would have been more than 30 percent higher in 2005. The authors also find that growth accelerations and decelerations have an asymmetric impact on human development outcomes. Finally, our results suggest that it is easier to identify the likely institutional and policy origins of growth decelerations than of growth accelerations.