|Author Name||ARATA Yoshiyuki (Fellow, RIETI) / MIYAKAWA Daisuke (Hitotsubashi University)|
|Creation Date/NO.||March 2022 22-E-027|
|Research Project||Determinants of Firm Dynamics: Causal Inference Approach|
|Download / Links|
Recent studies (e.g., Acemoglu et al (2012)) show that micro-level shocks propagate through an input-output network and result in aggregate fluctuations. While most previous studies report the propagation of shocks from upstream, we know little about how shocks propagate from downstream. Focusing on the sharp decline in exports from Japan during the global financial crisis and in consumption of food and accommodation services during the COVID-19 pandemic, we empirically examine how these demand shocks propagate from customers to suppliers through a firm-level input-output network. We find that the propagation of demand shocks depends on firm size and, in particular, on the mutual importance of the transaction relationship. For the case of the global financial crisis, negative demand shocks propagate from large exporters to larger suppliers because they regarded each other as their main transaction partners. In contrast, while small suppliers regarded these large exporters as their main partners, larger exporters do not see the small suppliers that way, and thus, the propagation via these transaction relationships is limited. For the case of the pandemic, negative demand shocks are transmitted from small customers even to their small suppliers. This is because many suppliers of firms that belong to pandemic-affected sectors are small, yet they are viewed as the main transaction partners by their customers. These results suggest that the mutual importance of transaction relationships determines the heterogeneity of the demand-shock propagation.