Networks and Business Cycles

The speed at which the US economy has recovered from recessions ranges from months to years. We propose a model incorporating innovation network, production network, and cross-sectional shock and show that their interactions jointly explain large variations in the recovery speed across recessions in the US. Besides the production linkages, firms learn insights on production from each other through the innovation network. We show that shock's sectoral distribution plays a crucial role in its amplification and persistence when the innovation network takes a low-rank structure. % Under a low-rank innovation network, the shock's impact on future growth is greatly amplified and persistent when the shock follows the sectors' importance vector in the innovation network. In contrast, the amplification becomes weak, and the shock's impact decreases exponentially when the shock follows other directions. We estimate a state-space model of the cross-sectional technology shock and document a set of new stylized facts on the structure of the innovation network and sectoral distribution of the shock for the US. We show that the specific low-rank network structure and the time-varying sectoral distribution of the shock can well explain the large variation in the recovery speed across recessions in the US. Finally, to emphasize the prevalence of the channel, we explore the application of the theory in asset pricing.