Works in Progress:
Innovation and Firm Growth
This paper sheds light on the post-entry dynamics of firms; namely that the mean, variance and skewness of the growth rate distribution is declining in age. I propose endogenous innovation as a mechanism which can capture these facts. Young firms take bigger risks in the innovation process and this, coupled with endogenous exit, leads to the post entry dynamics observed in firm level datasets. Using Compustat and NBER Patent data I present evidence in favour of my mechanism. In particular, I find that the size of the inventive step of a new discovery is decreasing in the age of the firm which patents the innovation.
Firm Productivity Over the Life Cycle: A Production Function Approach
In this paper I construct a measure of firm level productivity and trace out its evolution over the life cycle. I allow productivity to follow a Markov process which depends on age as well as spending on research and development. To do this I estimate a production function using the proxy variable methods of Olley and Pakes (1996). I find that age and research spending have similar effects on the interpretation of firm productivity growth. As well, I argue that controlling for endogenous exit is very important for interpreting productivity growth, in contrast to much of the literature. I argue that a model of firm growth must account for both the sluggish acquisition of factors of production as well as the declining variance of productivity growth.
The Decline in US Business Dynamism
This paper traces the impacts of a decline in the entry rate on aggregate labour markets and productivity using the framework of Restuccia and Rogerson (2010). A model of firm dynamics with convex hiring costs is calibrated to US firm level data. I examine the effects of distorting entry, exit and labor adjustment through the use of “wedges”. I find that the most plausible cause of declining business dynamism is a distortion to the entry margin and that the effect is felt primarily through a decline in aggregate productivity. The labour market effects are minimal. I also emphasize that a shift in the age distribution due to a decline in entry can explain a non-trivial portion of the pre-2000 decline in job creation. In this sense the decline in various measures of “dynamism” are partially due to one phenomena, the shift in the age distribution.