Microeconomic shocks and macroeconomic fluctuations in a dynamic network economy

The distribution of firm-sizes in the U.S. – or at least its upper tail – appears to be well-described by a Pareto distribution with infinite variance. This fact forms the basis of the granular hypothesis proposed by Gabaix in his paper “The Granular Origins of Aggregate Fluctuations” (Econometrica, (2011)). The granular hypothesis provides a mechanism whereby independent firm-level shocks are capable of generating macroeconomic fluctuations. This paper considers the granular hypothesis in a new framework. It develops a DSGE model by superimposing a stochastic overlapping generations framework on a network. Idiosyncratic output shocks to individual firms are transmitted across the economy through income–expenditure channels. Specifically, firms represent vertices of the network, and a firm x is linked to another firm y if x employs one or more workers who purchase commodities produced by y. The paper’s findings agree closely with results first discovered by Gabaix: if firm-sizes in an economy are described by a Pareto distribution, then independent firm-level shocks can generate macroeconomic fluctuations in accordance with the granular hypothesis. Furthermore, the model is capable of generating aggregate volatility of the same order of magnitude as occurs in reality. Thus the paper describes a new general equilibrium framework where macroeconomic fluctuations can arise as the consequence of independent firm-level shocks.