Global value chain participation and firm capacity utilization: Evidence from China

Abstract

Based on firm- and customs transaction-level data from 2000 to 2006, this study employs the translog cost function to measure the capacity utilization of Chinese firms. It then uses a difference-in-differences (DID) model combined with propensity score matching (PSM) to examine the impact of global value chain (GVC) participation on capacity utilization. The empirical results show that GVC participation has a significant positive impact on firm capacity utilization and that this impact lasts for a long time. Our results are robust after considering several potential problems. Mechanism testing reveals that GVC participation improves firm capacity utilization by expanding market demand and promoting technological progress. Heterogeneity analysis shows that GVC participation has a more significant impact on general trade firms, private firms, high-tech industry firms and firms in the western region of China. This study reveals that GVC participation can effectively ameliorate China's overcapacity, providing new ideas for achieving overcapacity governance.

KEYWORDS:

• GVC participation
• capacity utilization
• overcapacity governance

JEL Classifications:

• D24
• F14
• L60

Acknowledgments

We are grateful to the editor and anonymous reviewer for the comments and suggestions which led to an improved version of this paper.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 We also use the $\text{Probit}$ model to estimate propensity scores. Different estimation methods do not affect the robustness of the conclusions.

2 In this study, the BEC-HS1996 concordance table was adopted from 2000 to 2001, and the BEC-HS2002 concordance table was adopted from 2002 to 2006. The BEC codes of the intermediate inputs are "111", "121", "21", "22", "31", "322", "42", and "53".

3 Although we remove samples with missing or unreasonable key indicators in the data preprocessing stage, there are still some extreme values after the estimation of capacity utilization. To avoid the influence of extreme values, we truncate the capacity utilization by deleting samples with 0.25% of the upper and lower parts.

4 In this study, the balance and common support tests are carried out by year. Figure 6 presents the samples in 2001 as an example, and the other years also pass the balance and common support tests.

5 CEM coarsens each variable into substantively meaningful groups, performs exact matches on these coarsened data, and then retains only the original (uncoarsened) values of the matched data. CEM dominates commonly used existing matching methods in its ability to reduce imbalance, model dependence, estimation error, bias, variance, mean square error, and other criteria (Blackwell et al. 2009).

6 We follow Brandt et al. (2017) to compute output and input tariffs. The dataset of Chinese tariffs can be downloaded from the WTO website and the World Integrated Trade Solution website. FDI is measured by the number of foreign-invested firms. Soeshare is measured by the ratio of the number of SOEs to the number of domestic firms.

Additional information

Funding

The work was financial support from Zhejiang Provincial Philosophy and Social Sciences Planning Project [grant number 23NDJC017Z]; National Natural Science Foundation of China [grant numbers 42201227, 72073122]; and Fundamental Research Funds for the Provincial Universities of Zhejiang [Grant Number JR202202].