Inflation Diffusion Through Production Networks

This study investigates the inflation dynamics across industries using a dataset of South Korean manufacturing industries over 56 years, starting from 1965 to 2021. This paper aims to answer the following questions: (1) How does producer price inflation diffuse through the economy? (2) In which cases does the underlying production network explain the propagation of inflation? To answer these questions, we start by building both the production and inflation networks of the manufacturing sector. The inflation networks are constructed by relying on the VAR analysis that provides the forecast error variance decompositions. With this decomposition, we employ the Diebold-Yilmaz Connectedness methodology to transform these metrics into weighted and directed networks. Tracking the diffusion metrics presents how shifts in industrial policy targets and underlying input-output linkages are responsible for the inflation transmission. This study also shows the robustness of the analysis by extending the VAR decomposition to VARX to control for the fluctuations in the exchange rate and commodity prices