A simultaneous investigation of the environmental Kuznets curve for the agricultural and industrial sectors in China

Abstract

We investigate the relationship between China’s high growth and environmental pollution from the agricultural and industrial sectors by estimating the environmental Kuznets curve (EKC). We used panel data of Chinese cities at the prefecture level and above for the period between 2011 and 2015 and estimated the EKC models using the system generalized method of moments (GMM). We found that agricultural chemical oxygen demand (COD) and industrial wastewater depict N-shaped and reverse N-shaped curves. Further, an EKC for agricultural nitrogen balance has been established. We find that industrial sulfur dioxide (SO₂) emissions and their emission intensity support the EKC hypothesis. Our research is then used to provide policymakers with guidance on mitigating environmental pollution by comparing and scrutinizing the characteristics of agricultural and industrial pollution in China.

Keywords:

• Chinese prefecture-level city
• environmental Kuznets curve (EKC)
• environmental pollution
• system GMM
• turning point

Disclosure statement

No potential conflict of interest was reported by the authors.

Data availability statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Notes

1 We measure the value from a seven-year moving average calculated using data from IMF’s World Economic Outlook Database (https://www.imf.org/external/pubs/ft/weo/2018/02/weodata/index.aspx). The values for Japan, Taiwan, and South Korea are, respectively, 1.6%, 4.9%, and 5.3%.

2 We measure the value from a five-year moving average calculated using the data provided by the National Bureau of Statistics of China (http://data.stats.gov.cn/easyquery.htm?cn=C01).

3 Regarding the measurement method and data source, see note 2.

4 We obtain the data from World Development Indicators (http://databank.worldbank.org/data/source/world-development-indicators). The unit of energy consumption is indicated in kg of oil equivalent.

5 See note 4 about the data source.

6 Their air pollution data were from the Global Environmental Monitoring System (GEMS) project. Specifically, please refer to pages 8–11 in Grossman and Krueger (1991).

7 Per Grossman and Krueger (1995) table in Appendix 1 on page 373, parameters related to the GDP per capita in the estimated equation of SO₂ are significant.

8 The model is beneficial for reducing various types of econometric problems. Stern (2017) describes that those problems are generated from the variety of studied countries and spurious correlation of time series data by integrated variables.

9 Estimated parameters of the two-step model in system GMM and the dynamic fixed effects (DFF) and pooled mean group (PMG) model in panel ARDL method satisfy the sign condition of the EKC hypothesis in Table 2–Table 6 in Li, Wang, and Zhao (2016).

10 Proxies that represent external openness and the intensity of the production factor are, respectively, the ratio of foreign investment inflow and GDP and labor-capital ratio.

11 Of course, it is possible to extend the equation (4) adding other explanatory variables.

12 We refer to Blundell and Bond (1998) for the following explanatory part.

13 We apply the system GMM using the xtabond2 command in STATA with reference to Roodman (2009b). Utilizing other options of the command, Arellano and Bover (1995) orthogonal deviation transform is applied to minimize data loss in panels caused by eliminating the regional fixed effects. In addition, the collapsed instruments are used to control the increase of the instruments based on Roodman (2009a).

14 It is expected that the null hypothesis of no serial correlation is not significantly rejected in AR(2) tests.

15 It is expected that the null hypothesis of overidentifying restrictions is not significantly rejected. In addition to the results of the Hansen test, we also report the results of difference-in-Hansen test, including the test results of GMM and standard instruments for levels and differences based on Roodman (2009a). The results show only a summary. While the denominator indicates the number of all tests, the numerator means the number of tests that is not rejected at the 5% critical values.

Additional information

Funding

This work was supported by the Japan Society for the Promotion of Science (JSPS) under KAKENHI Grant Number 17K03698.

Notes on contributors

Shota Moriwaki

Shota Moriwaki is a Professor at Graduate School of Economics, Osaka City University, Japan. His research interest includes economic development in East Asian countries. He has published research articles in Asia-Pacific Journal of Regional Science and Asia & the Pacific Policy Studies.

Masayuki Shimizu

Masayuki Shimizu is an Associate Professor at Faculty of Global and Regional Studies, University of the Ryukyus, Japan. His research interest includes economic development in China. He has published research articles in Review of Development Economics and Journal of Chinese Economic and Business Studies.