"This paper presents an endogenous growth model that explains the evolution of the first andsecond moments of productivity growth at the aggregate and firm level during the post-war period.Growth is driven by the development of both (i) idiosyncratic R&D innovations and (ii) generalinnovations that can be freely adopted by many firms. Firm-level volatility is affected primarily bythe Schumpeterian dynamics associated with the development of R&D innovations. On the otherhand, the variance of aggregate productivity growth is determined mainly by the arrival rate ofgeneral innovations. Ceteris paribus, the share of resources spent on development of generalinnovations increases with the stability of the market share of the industry leader. As market sharesbecome less persistent, the model predicts an endogenous shift in the allocation of resources fromthe development of general innovations to the development of R&D innovations. This results in anincrease in R&D, an increase in firm-level volatility, and a decline in aggregate volatility. The effecton productivity growth is ambiguous.On the empirical side, this paper documents an upward trend in the instability of marketshares. It shows that firm volatility is positively associated with R&D spending, and that R&D isnegatively associated with the correlation of growth between sectors which leads to a decline inaggregate volatility"--National Bureau of Economic Research web site.