Effects of crude oil shocks on the PPI system based on variance decomposition network analysis

摘要

Previous studies have focused mainly on the price transmission from crude oil to producer price index (PPI). However, the PPI is an aggregate price index and consists of some producer price indices in different industrial sectors (sub-PPIs), Besides, there also exist transmissions relationships among sub-PPIs. Then, what are the price transmission mechanism from international crude oil to the sub-PPIs? Aiming to explore it, we propose a systematic framework combining the Granger causality test, the generalized variance decomposition method, and the network analysis method. The spot price of Brent crude oil and 15 Chinese sub-PPIs from January 1999 to November 2018 are taken as data samples. We find that (1) the full-sample analysis provides a useful summary of "average" transmission relationships, and the influence of crude oil to producer price volatility of petroleum-related industrial sectors is larger than others. (2) the rolling-sample analysis reveals that the empirical period can be divided into four phases, and there is a lag time in terms of the impact of crude oil price shocks on the transmission relationships between crude oil and sub-PPIs. (3) not all the sub-PPIs are affected by international crude oil price fluctuations in the empirical period, except for the petroleum industry.