Contemporaneous causality among price indices of ten major steel products

Abstract

The problem of dynamic relationships among the price indices of 10 major steel products – the rebar, steel wire, plate, hot rolled coil, cold rolled plate, galvannealed sheet, seamless tube, welded tube, section and narrow strip – is addressed in the present work for the Chinese market from 2011M7 to 2021M4. For examination of contemporaneous causal links among the 10 series, we use data on a daily basis and combine the vector error correction model and directed acyclic graph. This analysis is done using both the Peter and Clark and linear non-Gaussian acyclic model algorithms. With the exception of the price series for steel wire, each of the 10 series is part of cointegration relationships according to the vector error correction model, and all save the price series for thin strips respond to long-run equilibrium disturbances. The linear non-Gaussian acyclic model method allows us to achieve causal routes that allow for the examination of innovation accounting, but the Peter and Clark algorithm prevented us from reaching an acyclic graph. We categorise complex dynamics among price adjustment processes after shocks based on impulsive responses, for which the price indices of the plate, seamless tube and thin strip are predominating in comparison to the other seven items. Our findings show that these three goods should get the most consideration when designing long-term strategies for steel prices.