ABSTRACT
The domestic production network in China has been affected by multiple shocks in recent years, and this study reveals the dynamic changes of the shocks’ effects. Based on a domestic trade model with industry-level unobserved factors, we examine the trends of multiple shocks to the domestic market of China’s manufacturing industries and find that the domestic market has been bolstered by upward shocks until 2010, while has been depressed by downward shocks thereafter. Furthermore, based on multi-regional input–output analysis, we evaluate the impacts of domestic market shocks on industries’ value-added in the domestic value chain (DVC). We find that from 2004 to 2012, the domestic market shocks significantly promoted industries to participate in DVC embodied in the global value chain (DVC1), while after 2012 the shocks began to promote industries to participate in DVC completely based on endogenous capabilities (DVC2). However, the DVC2 position has not been upgraded by domestic market shocks. From another perspective, we investigate the impacts of domestic market shocks on the industrial linkage structure and find that the impact on industries’ centrality turns from negative to positive after the 2008 global financial crisis.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Supplementary material
Supplemental data for this article can be accessed online at https://doi.org/10.1080/00036846.2023.2210827
Notes
1 DVC is a chain of production in which the segments are all conducted in domestic regions.
2 More specifically, in the domestic production process, the raw materials and intermediate components may be procured at home or abroad, the final products may be sold at home or abroad. In DVC1, raw materials and intermediates are imported from foreign countries or the final products are exported to the external market, while in DVC2, all the segments are done locally.
3 GS = (Value Added Tax + Sales Tax +Income Tax)/Prime Operating Revenue..
4 It is notable that the covariance matrix of is a non-linear transformation of , and the correlation between will not in turn affect , thus there is no endogeneity problem in model (8)..
5 Specifically, the industry-level factor explains 48.19%, 34.08%, 38.74%, 36.04%, 56.36%, 56.31%, 74.02%, 37.72%, 61.39%, 45.00%, 46.56%, 32.08% and 35.29% of the variance for 13 industries, respectively.