Cross-Sectional Heterogeneity and the Persistence of Aggregate Fluctuations

Abstract

The micro evidence indicates that small firms grow faster than big firms. I argue that this relationship between the expected growth rate of a firm and its size may provide a micro foundation for the well-known high degree of persistence of shocks to aggregate output. The logic goes as follows. Almost any shock tends to temporarily alter firms' incentive to invest in growth thereby leading to a reallocation of firms across size categories. If small firms grow faster than big ones, the impact effect of the shock on aggregate output is gradually absorbed. But, as fast growing small firms become big and start to grow at the lower rate of big firms, the rate at which the shock is absorbed decreases over the adjustment path. As a result, shocks are absorbed, yet at a very low decreasing rate that induces long memory in aggregate output. I argue that this transmission mechanism may reconcile the micro evidence with the observed degree of aggregate persistence. It requires changes in neither the number of firms in the market nor the rate of technological progress. It is merely the result of the cross-sectional heterogeneity that we observe in real economies.

Keywords: Long memory, Gibrat's law, vintage models

JEL Classification: C43, E10, E32, L11