Pollution abatement efforts: a regional analysis of the Chinese industrial sector

Abstract

Although China has had rapid economic growth, it has borne a more significant economic burden or loss because of environmental pollution. However, the country has addressed this problem with various pollution abatement efforts. Some prior studies analysed the relationship between such efforts and pollution emissions, but did not show how these efforts affect pollution reduction. This study investigates the effects of pollution abatement efforts on industrial SO_X, NO_X and CO₂ emissions in the context of pollution reduction in China by panel data for 29 provinces from 1995 to 2010. The empirical results are as follows. First, emissions have increased rapidly in the 2000s. Second, rapid income growth has led to a greater increase in emissions. Third, pollution abatement would assist improvements in environmental quality. Further, this study reveals that abatement efforts affect emissions through the adoption of pollution removal measures.

Keywords:

• Chinese industrial sector
• emissions
• pollution abatement efforts
• pollution removal
• environmental Kuznets curve

Notes

1. Data from a poverty headcount figure of 1.90 dollars a day according to the 2016 world development indicators of the World Bank.

2. The three synchronisations relate to investment in new facilities which is to be synchronised with the design, construction and operation of appropriate waste treatment facilities (Ma and Ortolano 2000).

3. CEY (2000).

4. China Statistical Yearbook on Environment (CSYE 2006, 2011).

5. CSY (2011). The total expenditure includes investment in urban environmental infrastructure and the treatment of industrial pollution, and ‘three synchronisations’ environmental investment for new projects.

6. Only Brajer, Mead, and Xiao (2008, 2011) and Roumasset, Burnett, and Wang (2008) used data about pollution concentrations instead of pollution emissions.

7. As Du, Wei, and Cai (2012) pointed out, there were some cases wherein the inflection point of EKC, at which income continues to increase but pollution levels begin to decrease, is much higher than practical income levels. In addition, Song, Zhang, and Wang (2013) calculated the inflection point of income in each province and indicated that some provinces, such as Liaoning, Anhui, Fujian, Hainan and Qinghai, have no inflection point.

8. Liu (2012) defined environmental poverty as the lack of a healthy environment which is needed for society's survival and development.

9. In EKC studies, many simultaneously analysed other pollution-determining factors such as trade openness (Shen 2008; He 2010; Jayanthakumaran and Liu 2012), foreign direct investment (He 2006), income inequality (Zhang and Zhao 2014), et cetera.

10. He and Wang (2012) used a ratio of government environmental staff to the total number of governmental staff as a proxy variable of environmental regulation.

11. Yin, Zheng, and Chen (2015) used a ratio of industrial pollution abatement expenditure to regional GDP as a proxy variable of environmental regulation.

12. Shaw et al. (2010) introduced an environmental policy variable which is the cumulative number of years for which a city has been designated as a key environmental protection city.

13. According to CEY (2012), China’s cost of desulphurisation and denitration accounted for 59% of pollution abatement expenditure on industrial waste gas in 2011. The breakdown is 53% for the cost of desulphurisation and 6% for the cost of denitration. The former is clearly much higher than the latter.

14. Tibet is excluded because the data are unavailable. Likewise, the data are unavailable for Hainan in 2002 and for Ningxia from 2000 to 2002. Sichuan is included in Chongqing because the data for Chongqing are available from 1997.

15. Using NISTEP (1992) has the following advantage: it has the emission factors of SO_X, NO_X and CO₂ in China which are not published in official statistics such as the CEY. Further, the emission factors are taken from the results of prior researches and field surveys; thus, they reflect regional and industrial characteristics in China. See Appendix 1.

16. In general, the energy balance table is described as the energy consumption statistics divided between ET and FC. In this study, ET is included in TPG, heating supply, coking, petroleum refineries and gas works. FC is included only in the industry sector. See Appendix 1.

17. The CESY is divided basically into 17 fuels as follows: coal (which is divided into raw coal, cleaned coal, other washed coal and briquettes), coke (which is divided into coke, coke oven gas, other gas and other coking products), petroleum (which is divided into crude oil, gasoline, kerosene, diesel oil, fuel oil, liquefied petroleum gas, refinery gas and other petroleum products) and natural gas. See Appendix 1. Since 2010, however, the number of fuels has increased from 17 to 27 in the CESY.

18. Appendix 2 shows the sulphur contents of coal by province and petroleum for each product in China.

19. See Appendix 3. In this regard, the estimations of SO_X and NO_X emissions also convert fuel consumption to lower calorific values for some fuel types in Appendix 1.

20. According to CESY (2011), China’s coal consumption accounted for 68% of total energy consumption in 2010.

21. Recently, China imposed an obligation to reduce energy consumption per GDP unit by 20% compared with the 2005 level in the Eleventh Five-year Plan period, a level which CSY (2011) reported was achieved.

22. As described above, the removals data are also not covered by the entire sample. However, this study uses the data as it is, since it would appear that pollution removal measures have made little progress in the TVEs. Further, with regard to SO_X removals, the CEY published data about the volume of industrial SO₂ that has been removed.

23. Minami and Makino (2014) estimated the total population based on a population census of China. Hence, their estimates are different from the official CSY figures.

24. Burnside, Eichenbaum, and Rebelo (1995) showed that electricity consumption is a better proxy of capital stock including capital utilisation rates.

25. This study calculated the composition ratio of the industrial sector by province using the population census for the base years of 1990, 2000 and 2010. The results were then extrapolated from the base years to other years.

26. In order to select the fixed or random effect model, this study used the approach according to Arellano (1993) and Wooldridge (2002).

27. The inflection point in the level of GDP per capita was computed as [exp(−β₁/2β₂)] and in yuan at 1995 constant prices.

28. The inflection point of pollution abatement expenditure was computed as [exp(−β₃/2β₄)] and in ten thousand yuan at 1995 constant prices.