This study analyses the underlying factors of recently declining air pollution in China, utilizing annual time series data spanning from 2000 to 2020 in the framework of the autoregressive distributed lag model approach. In the empirical setting, air pollution is represented by particulate matter 2.5 concentrations known as the most detrimental ambient pollutant, and the empirical model includes several socioeconomic potential determinants of air pollution. The main motivation behind the study is the downward tendency of air pollution in China as of the second decade of the 2000s. At this point, although it is commonly accepted that the Air Pollution Prevention and Control Action Plan implemented by the Chinese State Council has been effective, the underlying specific factors of declining air pollution are not clear. From this motivation, the study unveiled the positive impact of economic growth, industrialization, and foreign direct investment inflows on air pollution, while medium- and high-tech exports and coal rents are found negatively associated with air pollution. In this respect, overall results particularly emphasize the declining impact of an increase in medium- and high-tech exports on air pollution. In this context, to improve the air quality further, the study suggests that China should transform its industrialization structure toward specialization in medium- and high-tech products, and promote foreign direct investment inflows specialized in these types of products. The study provides additional policy recommendations for Chinese policymakers.
Research Article
An Empirical Investigation on the Underlying Factors of Recently Declining Air Pollution in China
https://doi.org/10.21203/rs.3.rs-4006772/v2
This work is licensed under a CC BY 4.0 License
Version 2
posted
You are reading this latest preprint version
Air Pollution
Industrialization
Foreign Direct Investment
Medium-and-High-Tech Exports
China
Environmental issues have been increasingly dealt with globally, in particular as sustainable development goals propounded by the United Nations, and many countries have been striving to cope with environmental problems in their development paths. As a crucial environmental indicator, particulate matter 2.5 (hereafter PM2.5) concentrations, which are defined as the amount of fine particulate aerosol particles up to 2.5 microns in diameter, are the most detrimental air pollutants owing to their environmental prevalence and the wide range of adverse human health impacts associated with exposure. PM2.5 concentrations are measured in micrograms per cubic meter (µg/m3) and consist of sulfates, nitrates, black carbon, and ammonium. The World Health Organization (WHO) set the particle matter value at 10 µg/m3 as a guide indicator in 2005, yet updated this figure to 5 µg/m3 in 2021. PM2.5 concentrations exceeding the guideline value lead to severe health risks, and exposure to them is more dangerous for the elderly, children, asthmatics, and those with austere health conditions such as cancer and diabetes. In addition, they have long-term effects on the respiratory and circulatory systems, resulting in heart disease and deteriorated lung function (IQAir, 2022). UNEP (2023) reported that around four million people worldwide died from exposure to PM2.5 concentrations in 2019. IQAir (2022) classifies PM2.5 concentrations in nations by density per cubic meter in a color-coded chart[1]. All these PM2.5 statistics provide significant insight into the nexus between air pollution caused by PM2.5 concentrations and economic development levels. Consistent with this insight, the report demonstrates more than 90% of deaths from air pollution occur in low-and-middle-income countries such as African, Central Asian, and South Asian nations. As a most likely paradoxical situation for the air pollution-income level nexus, a significant statistic from UNEP (2023) indicates that China was responsible for approximately 35% of PM2.5-related deaths worldwide in 2019.
As of the economic reforms that began in 1978, China has achieved extraordinary economic expansion (Jiang et al., 2018; Hao et al., 2019; Xu et al., 2019). According to the World Development Indicators of the World Bank (2023), following these reforms, China performed notable annual growth rates in both aggregate and per capita levels. Over the period 1978-2021, China grew at a rate of 9.25%, on an annual average; on the other side, this statistic was at 2.99% and 2.37% for the world and Organization for Economic Cooperation Development (OECD) countries, respectively. Throughout this period, China showed much better growth performances than the world and the OECD averages, even during the periods of COVID-19 that emerged in China. While the world and OECD were experiencing negative annual growth rates right after the 2008-09 global financial crisis and COVID-19, China grew positively. This spectacular economic expansion; however, has caused severe environmental pollution problems, particularly PM2.5-based air pollution, which was measured at 58.5 µg/m3 in China in 2013 (more than 11 times of threshold value recommended by the WHO). In the same year, heavy smog was considerably observed in the Northern and Central areas of China, and concerns related to public health began to go up (Jiang et al., 2018; Cheng et al., 2023).
Spreading smog waves resulting from PM2.5 concentrations called for the Chinese government to intervene and take measures against the intensified air pollution. Eventually, in 2013, the Chinese State Council put into force the Air Pollution Prevention and Control Action Plan (APPCAP), which aims to promote manufacturing technology in environmentally friendly industries through research and development, innovation, and the implementation of new technologies (Zhou and Tang, 2021). The APPCAP primarily controls air pollution by (i) rewarding firms with high-energy efficiency and low emissions, (ii) providing business income tax incentives to high-tech companies with environmental protection projects, and (iii) offering financial support to air pollution control projects (Cheng et al., 2023). Herewith, the APPCAP mapped out an ecologically friendly development route through technology and innovation. Among others, Huang et al. (2018), Feng et al. (2019), and Zhang et al. (2019) stated that APPCAP improved overall air quality and lowered mortality caused by PM2.5 concentrations in China. On the other hand, Ding et al. (2019) estimated the contribution of APPCAP in reducing PM2.5 concentrations from 2013 to 2017 at the rate of 88.7%. Additionally, some studies highlighted that energy transitions from fossil fuels to clean fuels have significantly improved the air quality in China (see, Zhao et al., 2018; Zhu et al., 2019; Shen et al., 2021; Akça, 2024). Over the period 2019-2023, China incomparably takes the lead in increasing clean energy investment among all countries, with an increase of almost 185 billion US dollars, even more than that of the European Union (International Energy Agency, 2023). As well as the APPCAP and energy transition, other factors possibly led to dropping in China’s PM2.5 concentrations from 58.5 µg/m3 to 34.8 µg/m3 (from maroon to red)[2] over the period 2013-2020 (OECD, 2023). Nevertheless, despite all air pollution preventive practices, the level of PM2.5 concentrations in China is still quite higher than the WHO guideline indicator.
Drawing attention to socioeconomic factors of the air pollution problem caused by PM2.5 concentrations in China, IQAir (2022) underlines that the use of coal and other non-renewable energy sources is still pervasive in China. The report also enlightens us on China's prominent structural characteristics, such as being a rapidly industrializing nation with abundant natural resource reserves, industrial-oriented production and trade, and a large population, which all possibly cause high PM2.5 concentrations. Numerous empirical studies have analyzed several socioeconomic factors, such as population density, urbanization, economic growth, industrial activity, trade openness, and natural resource abundance, to identify probable predictors of PM2.5 concentrations in China. Investigating the associations between air quality and socioeconomic factors within the framework of hypothesis-based approaches is crucial for more effective policy recommendations. In this context, empirical studies have tested the Environmental Kuznets Curve (EKC) hypothesis, expressing that, at the early stages of development, environmental degradation increases as the economy grows but environmental recovery begins as growth reaches a threshold value (Grossman and Krueger, 1996). Testing the EKC hypothesis for the rapidly expanding Chinese economy significantly advances the grasp of knowledge about the association between environmental quality and its determinants. Besides, the resource curse hypothesis put forward by Auty (1993) contends that natural resource abundance negatively affects economic development through different channels. One of these channels may be related to environmental issues, and herewith this hypothesis might offer fundamental information on environmental degradation in a specific period for a natural resource-rich and rapidly growing developing country like China. In addition, there are hypotheses based on foreign direct investment (FDI) inflows in countries that have grown with the significant contribution of FDI inflows, such as China. Contrasting approaches, the pollution halo hypothesis, and the pollution heaven hypothesis analyze the environmental effects of FDI inflows. Supported by Talukdar and Meisner (2001), Wheeler (2001), and Perkins and Neumayer (2008), the pollution halo hypothesis claims that FDI inflows bring about lower carbon emissions and, consequently, emphasizes the qualified and environmentally friendly nature of FDI inflows. In contrast, empirically supported by Grimes and Kentor (2003), Cole (2004), Hoffmann et al. (2005), and Jorgenson (2007), the pollution heaven hypothesis reveals that FDI inflows increase the carbon emission level, and consequently it underlines the labor-intensive and polluting character of capital imported from abroad.
The economic development path with its dynamics and the course of PM2.5 concentrations provides us with crucial sources of motivation to scrutinize the PM2.5-related air pollution problem in China in conjunction with its level of economic development. From this motivation, this study discloses the primary factors contributing to considerably moderating the air pollution problem caused by PM2.5 concentrations in China, utilizing appropriate econometric methods. The study's contribution to the literature is as follows: (i) The best-fitting econometric techniques as per the data features are employed to provide robust results. (ii) The study yields evidence regarding the validity of the natural resource curse hypothesis from an environmental perspective for China (iii) The study also hierarchically tests the pollution haven or pollution halo hypotheses for China (iv) The study highlights the significance of medium- and high-tech exports in lowering PM2.5 concentrations through strengthening China's technological transformation in foreign trade. (v) Thus, based on the empirical findings, our study may also guide the policy-making process of nations with severe air pollution problems to improve air quality by underlying the reducing factors of PM2.5 concentrations in China. The remainder of the study is structured as follows: The primary empirical studies, which examined the nexus between socioeconomic factors and air quality in China, are thoroughly reviewed in Section 2. The econometric model and data employed in the empirical analysis are described in Section 3. The analysis results and discussion are included in Section 4, and Section 5 mentions the concluding remarks.
[1]According to this color-coded ranking based on the intensities of PM2.5 concentrations, health risks increase as the color changes from blue to maroon. In this color-coded chart: blue 0-5 µg/m3, green 5.1-10 µg/m3, yellow 10.1-15 µg/m3, orange 15.1-25 µg/m3, red 25.1-35 µg/m3, purple 35.1-50 µg/m3, maroon over 50 µg/m3 (IQAir, 2022).
[2] Thanks to APPCAP, China outperformed high-income Middle Eastern countries such as Saudi Arabia (41.50 µg/m3), United Arab Emirates (45.90 µg/m3), and Qatar (42.50 µg/m3), African countries such as Chad (89.70 µg/m3) and Egypt (46.50 µg/m3), and finally Asian countries such as Pakistan (70.90 µg/m3), Bangladesh (65.80 µg/m3) and India (53.30 µg/m3) in reducing PM2.5 concentrations (IQAir, 2022).
In recent times, China has been one of the most researched countries, most likely taking the lead, in the context of the issues related to environmental quality since it has been intensively exposed to environmental problems in its economic development process. While a great majority of the studies in the empirical literature took an interest in the primary dynamics of environmental issues, carbon dioxide (CO2) emissions, in this vein, were commonly utilized as an environmental quality indicator. In one of them, Halkos and Tzeremes (2011) revealed an inverted U-shaped relationship between CO2 emissions and economic growth in China from 1960 to 2006, proving the validity of the EKC hypothesis. Reaching a similar finding, Ren et al. (2014) also established that China's rising trade surplus and FDI inflows are responsible for the sharp increase in CO2 emissions from 2000 to 2010. Extending this period to five years back (1995-2010), Zhang and Zhou (2016), conversely, detected the lowering impact of FDI on CO2 emissions for China, referring to the validity of the pollution halo hypothesis. Additionally, this positive impact of FDI inflows on environmental quality weakens from the Western region to the Eastern and Central regions.
In the context of the sustainable development goals put forward by the United Nations, which highlight the transition from non-renewable energy sources to renewable ones for a cleaner and more sustainable environment, China’s energy sources are mostly situated in the empirical literature exploring the primary dynamics of environmental quality. Analyzed the panel data spanning from 2001 to 2015 for 30 Chinese provinces, Lin and Zhu (2019) affirmed the cutting down function of renewable energy technological innovation on CO2 emissions. Likewise, Ahmad et al. (2020) found evidence that renewable energy sources in five Northwest Chinese provinces mitigate environmental pollution with different extends in both short and long terms from 1995 through 2014; on the other hand, the expansions in non-renewable energy sources, whole economic activity and transportation sector are associated with enhancing environmental pollution proxied by CO2 emissions. According to the results of Shen et al. (2021), while energy consumption, financial development, and natural resources rent increase CO2 emissions, green investments decrease them in 30 Chinese provinces over the period 1995-2017. Handled the same period (1995-2017), Guo et al. (2021) found evidence that income, environmental innovation, investment in the energy industry, and renewable energy consumption are critical factors that help explain CO2 emissions in China. The findings of the study also supported the validity of the EKC hypothesis. Hsu et al. (2021) uncovered that contrary to globalization; ecological innovation, renewable energy, and environmental taxes contributed to the reduction of CO2 emissions in China for the period 1980–2018.
Undoubtedly, CO2 emissions are commonly considered in empirical literature related to the nexus between socioeconomic factors and environmental quality owing to their harmful effects on human health. In this respect, PM2.5 concentrations are presumably among the worthiest considerations because of their highly detrimental impact on human well-being. In this context, recently, the number of studies examining the socio-economic determinants of PM2.5 concentrations has been more and more increasing. Worldwide, China is among the most taken into consideration countries from the point of view of air pollution based on PM2.5 concentrations due to its high degree of exposure to them. Hence, it is observed that either time series analysis regarding PM2.5 concentrations predominantly focused on China or panel data analysis includes China in their empirical setting. Besides, decentralized complex state systems in China typically pave the way for disparate economic conditions and sectors in miscellaneous regions of China. Within this framework, the empirical literature studying the dynamics of PM2.5 concentrations in China also takes into consideration spatial and time variations. Han et al. (2016), for instance, detected that population size and economic growth are the most significant determinants of PM2.5 concentrations using time series of China from 1973 to 2013. On the other hand, Li et al. (2016), by utilizing panel data from relatively stable prefecture-level cities, referred to urbanization, industrialization, and economic growth as underlying factors of accelerating PM2.5 concentrations in a very short time (1999-2001). Analyzing the determinants of PM2.5 concentrations in 29 Chinese provinces over the period 2001–2012, Xu et al. (2016) illustrated an association resembling an inverted U-shaped curve between economic growth and PM2.5 concentrations, which is consistent with the arguments of the EKC hypothesis. A similar association is also ascertained for the nexus of urbanization, coal consumption, and private vehicles with PM2.5 concentrations. On the contrary, the improvements in energy efficiency are associated with PM2.5 concentrations in a positive U-shaped pattern.
Employing data from 2007 on both exports and domestic trade, Wang et al. (2017) investigated PM2.5 concentrations and their effects on public health in China. According to the results, contrary to domestic trade, which contributes to better air quality in developed coastal provinces, 50% to 60% of China's air pollution stems from goods and services imported from other provinces. While PM2.5 concentrations give rise to almost 1.10 million premature deaths of which 19% are associated with exports, the health burden of air pollution in China's less developed inner provinces is deepened by both exports and interprovincial trade. Questioned the primary socioeconomic factors influencing PM2.5 concentrations in China for the period 2001–2012, the findings of Cheng et al. (2017) indicated an inverse U-shaped association between PM2.5 concentrations and economic growth, supporting the EKC hypothesis. Additionally, factors such as population growth, a high rate of secondary industries, a coal-dominated energy structure, and traffic have contributed to China's ascending air pollution problem. Employing 2010 data from 30 provinces of China, Wang et al. (2018) reached evidence that the trade of intermediates from China's interior to coastal areas contributes to additional PM2.5-related emissions. By contrast, trading of final products from coastal to inland results in a downfall in PM2.5-related emissions, corresponding with the pollution halo hypothesis. Zhao et al. (2018) unveiled that the largest contributor to the decrease in PM2.5 concentrations in China from 2005 through 2015 was the reduction of household fuel consumption despite trivial contributions from power plants, industry, and transport. On the other side, employing data from 190 Chinese cities in 2014, Zhou et al. (2018) showed that industrial structure, population density, industrial soot emissions, and road density have a remarkably positive impact on PM2.5 concentrations. In addition, the study revealed that trade openness and electricity consumption do not affect PM2.5 concentrations.
Reaching the findings supporting both the EKC and the pollution haven hypothesis for China over the period 1998–2015, Liu et al. (2019) proved that FDI inflows, industry structure, and road density are remarkable encouraging factors of PM2.5 concentrations, albeit population density does not significantly affect PM2.5 concentrations. Analyzing the contribution of rural housing emissions to PM2.5 concentrations and their impact on health and climate in China for the years 1992 and 2012, Shen et al. (2019) obtained results indicating that clean energy transitions significantly led to the reduction in PM2.5 concentrations, and herewith preventing nearly130,000 premature deaths. Similarly, Zhu et al. (2019) showed that rapid energy transitions and urbanization resulted in a nearly 50% reduction in residential energy consumption, mitigating considerably air pollutant emissions, in China from 1992 to 2012. Analyzing data from 281 prefecture-level cities in China over the period 2007–2017, Xu et al. (2021) found that PM2.5 concentrations have a spatial correlation among provinces in China, and industrial restructuring and technological progress mitigate PM2.5 concentrations. However, interestingly the interaction between industrial restructuring and technological progress causes the rise in PM2.5 concentrations. Zeduo et al. (2022) found a positive relationship between PM2.5 concentrations and urbanization, which resembles an N-type EKC association, in the Yellow River Ecological and Economic Region of China over the period 2010-2019. Exploring the PM2.5 concentrations in 269 Chinese cities from 2004 to 2016, Yan et al. (2022) discovered that PM2.5 concentrations led to an increment in the spatial associations among these cities, and this increment was more sharply observed in the cities located in Northern China. The study also discovered that at the most, population density contributed to urban PM2.5 concentrations and this contribution was homogeneous in time. On the contrary, economic output, secondary industry, and FDI have heterogeneous impacts on PM2.5 concentrations in time and their impacts may be overestimated unless considering the neighborhood among cities. Wang and Cao (2023), for the period 2005–2020, ascertained the negative association of PM2.5 concentrations with general public budget revenue despite their positive association with economic output, secondary industry, and population density in 359 Chinese cities. Following a different strategy from these studies and including China in a wide range of panel groups of 60 countries over the period 2010-2019, Akça (2024) highlighted the efficient role of renewable energy sources in improving air quality through mitigating PM2.5 concentrations.
In addition to CO2 emissions and PM2.5 concentrations, some studies focused on other air quality indicators in China, such as sulfur dioxide and nitrogen oxides. Moreover, studies exploring the nexus between socioeconomic factors and air quality in China also employed factors such as green total factor productivity, ecological efficiency, and natural gas use. In one of them, Wang and Chen (2014) examined sulfur dioxide emissions at the national and regional levels in China for the period 2002–2009. The findings proved that while FDI inflows from OECD nations strengthen emissions, FDI inflows from Hong Kong, Macau, and Taiwan have no discernible impact on emissions. Besides, the study uncovered that institutional improvement weakens the overall polluting effects of FDI inflows. Jiang (2015) explored the effects of regional FDI inflows openness and regional economic growth on polluting emissions in 28 Chinese provincial-level areas using panel data from 1997 through 2012 and indicated that FDI inflows are responsible for major polluting emissions. The study also confirmed the validity of the EKC pattern. Xu et al. (2019) handled sulfur dioxide, nitrogen dioxide, and PM2.5 concentrations in China between 2005 and 2015. The results of the study indicated that as well as an inverted U-shaped relationship between urbanization and air pollutants in all regions, economic growth reduces air pollutants in the eastern region of China while increasing in other regions. Population growth and urbanization promote sulfur dioxide and nitrogen dioxide, despite the reducing PM2.5 concentrations in eastern China. Whereas, they reduce all air pollutants in the central and western regions of China. Lastly, energy efficiency is found to be an encouraging factor for air quality. Analyzing the panel data of 30 provinces in China from 2003 to 2016 in a different window, Cheng et al. (2020) detected that the cities where there are abundant natural resources have less level of green total factor productivity, and hence natural resource abundance significantly hinders China's green growth.
The study builds an econometric model to capture the underlying dynamics of PM2.5 concentrations utilizing China’s annual time series spanning from 2000 to 2020. As it is understood from the empirical literature review, several factors could be incorporated into the econometric model; yet data availability and analysis framework constrict the variable choice, particularly given that PM2.5 concentration series are consecutively started from the year 2000. However, after taking a glance at China's socioeconomic dynamics and transformation that happened as of the beginning of the 21st century, the most notable variables are included in the econometric model. These variables also have been argued on a global scale in recent times in the context of determinants of air quality. Herewith, the study estimates the following dynamic stochastic regression model:
In Eq. 1, ln refers to the natural logarithmic operator and ε is the composite error term. Air pollution indicator PM2.5 is the dependent variable that shows mean population exposure to PM2.5 concentrations in China from 2000 through 2020 and data for this variable is retrieved from the OECD (2023) database. The econometric model of the study involves seven independent variables thought to be conducive to more or fewer PM2.5 concentrations in China. The first independent variable is gross domestic product per capita (GDPpc) at constant 2015 prices and shows the economic growth path in US dollars. The second one IND represents the industrialization level and is measured at the value-added share of industry including construction in GDP. Another variable is foreign direct investments (FDI) measured as a share of net inflows in GDP. Medium and high-tech exports (MHTE) is the share of medium and high-tech manufactured exports in total manufactured exports and renewable energy consumption (REC) is the share of renewable energy in total final energy consumption. Coal rents (CR) are the difference between the value of both hard and soft coal production at world prices and their total costs of production and expressed as a percent of GDP. Lastly, research and development expenditure (RD) shows the gross domestic expenditures on research and development expressed as a percent of GDP. All the independent variables are retrieved from the World Bank’s (2023) World Development Indicators. In Table 1, descriptive statistics of all variables and the Pairwise Pearson correlation matrix are demonstrated.
Table 1 Descriptive statistics and Pairwise correlations
|
Descriptive statistics (non-logarithmic) |
|||||||||
|
Variables |
Mean |
Median |
Std. dev. |
Max. |
Min. |
||||
|
PM2.5 |
49.123 |
50.861 |
5.651 |
58.481 |
34.841 |
||||
|
GDPpc |
5858.151 |
5647.069 |
2720.203 |
10358.17 |
2193.897 |
||||
|
IND |
43.943 |
45.223 |
3.164 |
47.557 |
37.843 |
||||
|
FDI |
3.014 |
3.475 |
1.052 |
4.554 |
1.311 |
||||
|
MHTE |
57.103 |
58.344 |
4.329 |
61.365 |
45.491 |
||||
|
REC |
16.401 |
14.141 |
5.858 |
29.631 |
11.341 |
||||
|
CR |
1.383 |
0.802 |
1.259 |
4.953 |
0.147 |
||||
|
RD |
1.661 |
1.714 |
0.464 |
2.406 |
0.893 |
||||
|
Pairwise correlations between variables (logarithmic) |
|||||||||
|
Variables |
PM2.5 |
GDPpc |
IND |
FDI |
MHTE |
REC |
CR |
RD |
|
|
PM2.5 |
0.00 |
|
|
|
|
|
|
|
|
|
GDPpc |
-0.53 |
0.00 |
|
|
|
|
|
|
|
|
IND |
0.75 |
-0.68 |
0.00 |
|
|
|
|
|
|
|
FDI |
0.66 |
-0.72 |
0.94 |
0.00 |
|
|
|
|
|
|
MHTE |
-0.33 |
0.76 |
-0.31 |
-0.38 |
0.00 |
|
|
|
|
|
REC |
0.11 |
-0.84 |
0.25 |
0.36 |
-0.87 |
0.00 |
|
|
|
|
CR |
0.31 |
0.02 |
0.58 |
0.47 |
0.43 |
-0.45 |
0.00 |
|
|
|
RD |
-0.48 |
0.98 |
-0.67 |
-0.71 |
0.83 |
-0.88 |
0.06 |
0.00 |
|
According to the descriptive statistics, the mean of PM2.5 concentrations is 49.123 µg/m3, of which the maximum value was 58.481 in 2013 and the minimum value was 34.841 in 2020 within the research period. China paints a different image in the first and second decades of the 21st century in terms of PM2.5 concentrations. While the first decades testified to a modestly rising trend of PM2.5 concentrations, in the second decades, especially as of 2013, China made substantial progress in reducing PM2.5 concentrations. The GDP per capita of the Chinese economy is at a value of 5 thousand 858 dollars, on average, over the period 2000-2020, together with a continuously increasing trend, and reached the value of almost 10 thousand 358 dollars in 2020. On the other hand, the mostly descending trend of industrialization may be interpreted as a change in the components of economic growth. The share of value-added of industry in GDP was, on average, at nearly 44%, and dropped to the level of 37.8% in 2020. A similar situation is observed in net FDI inflows, which consist of nearly 3% of GDP, on average, from 2000 to 2020. While net FDI inflows corresponded to 4.5% of GDP in 2005, their share in GDP dropped to 1.3% in 2019. On the contrary, medium and high-tech exports have an increasing trend in the relevant period. They correspond to 57% of total manufacturing exports, on average, with a maximum value of 61.4% in 2020 and a minimum value of 45.5% in 2000. Another salient statistic is about renewable energy consumption, of which the mean value is 16.4% during the research period, even if the absolute value of renewable energy consumption in China has gone up, its share in total final energy consumption remarkably went down compared to that of the beginning of 2000s. Having a quite unsteady course, coal rents, on average, consist of nearly 1.4% of GDP, together with a maximum value of about 5% in 2008. Lastly, research and development expenditure has gradually increased from 2000 to 2020 and corresponds to 2.4% of GDP in 2020. Figure 1 presents the course of the variables, which are scaled by the 2010=100 index, to just observe the path of the variables across the pertinent period. Accordingly, as of 2013, a continuous downfall in PM2.5 concentrations is clear, and coal rents are generally quite unstable. Additionally, while GDP per capita consistently soared, the share of renewable energy consumption in total final energy considerably scaled down.
The results of pairwise correlations matter especially in detecting multicollinearity problems among independent variables. In this respect, there is a high correlation between GDPpc, REC, and RD. Similarly, FDI is highly correlated with IND, as well as the high correlations among MHTE, REC, and RD. PM2.5 concentrations are positively correlated with IND, FDI, REC, and CR; on the other hand, they are negatively correlated with GDPpc, MHTE, and RD. Following these preliminary statistical analyses, the strong proof regarding the underlying factors of PM2.5 concentrations needs necessary econometric analysis.
Analyzing the underlying factors of PM2.5 concentrations in China, this study follows the procedure of time series analysis. In this context, firstly here some unit root tests on the variables are conducted since that matters for determining the most fitting analysis techniques. Without detecting the stationary situations of the variables, getting unbiased and efficient estimation results may not most likely be warranted. For instance, in the case of non-stationary variables, estimations of ordinary least squares have been spurious, and cannot yield reliable results. Again, econometric models, which involve variables with different stationary situations, require a method peculiar to this model characteristic. Hence, before passing to an estimation of the econometric model in Eq. 1, the most commonly used unit root tests in the literature Augmented Dickey-Fuller (ADF) and Philips-Perron (PP) tests are performed to determine the integration degrees of all series and the results are presented in Table 2.
Table 2 Unit Root Test Results
|
Variables |
|
ADF |
PP |
||
|
Test stat. |
Prob. |
Test stat. |
Prob. |
||
|
lnPM2.5 |
Level |
0.006[0] |
0.948 |
0.167[2] |
0.963 |
|
First Difference |
-5.445[0]* |
0.000 |
-5.328[2]* |
0.000 |
|
|
lnGDPpc |
Level |
-3.915[0]* |
0.001 |
-3.023[2]** |
0.049 |
|
First Difference |
-- |
-- |
-- |
-- |
|
|
lnIND |
Level |
-0.906[0] |
0.935 |
-0.985[1] |
0.924 |
|
First Difference |
-3.494[0]*** |
0.068 |
-3.468[1]*** |
0.072 |
|
|
lnFDI |
Level |
-0.906[0] |
0.765 |
-0.792[2] |
0.799 |
|
First Difference |
-5.151[0]* |
0.001 |
-5.209[2]* |
0.000 |
|
|
lnMHTE |
Level |
-4.817[0]* |
0.001 |
-4.619[2]* |
0.002 |
|
First Difference |
-- |
-- |
-- |
-- |
|
|
lnREC |
Level |
-2.947[0]*** |
0.058 |
-2.653[2]*** |
0.099 |
|
First Difference |
-- |
-- |
-- |
-- |
|
|
lnCR |
Level |
-2.514[0] |
0.127 |
-2.487[2] |
0.133 |
|
First Difference |
-6.024[0]* |
0.001 |
-6.182[2]* |
0.001 |
|
|
lnRD |
Level |
-2.393[0] |
0.156 |
-2.393[2] |
0.156 |
|
First Difference |
-3.312[0]** |
0.028 |
-3.312[2]** |
0.029 |
|
Note: *, **, and *** mean that test statistics are significant at a 1%, 5%, and 10% levels, respectively. The values in brackets show the optimal lag length determined according to the Schwarz Information Criteria for ADF tests and show the bandwidth determined according to the Barnet Kernel predictor for the PP test.
According to both unit root test results, the variables of GDPpc and REC are stationary at levels, and other variables have been stationary by taking their first differences, i.e., all variables are not integrated in the same order. These stationary characteristics of the variables cannot enable us to perform the ordinary least squares estimator, but the existence of a level relationship between the dependent variable and a set of regressors can be questioned by employing a variety of cointegration methods. In our case, we should apply for the autoregressive distributed lag (ARDL) bounds testing approach originated by Pesaran et al. (2001) owing to the series being integrated at different orders. The bounds testing approach is based on standard F- and t-statistics, of which asymptotic distributions are non-standard under the null hypothesis referring to no level relationship between variables irrespective of whether the regressors are I (0) or I (1), by which significance of the lagged levels of the variables in a univariate equilibrium correction mechanism is tested.
The ARDL cointegration approach incorporates several advantages compared to other cointegration techniques (Odhiambo, 2009): First and the most important one is that there is no restrictive assumption on the variables like they must be integrated in the same order. That is to say, the ARDL approach can be applied without considering whether the series are integrated of order zero or order one. However, this feature of the ARDL approach does not remove the need for performing the unit root tests because some series may be stationary at second order, even if the possibility is quite low. Secondly, the ARDL approach is not sensitive to the sample size as much as other techniques, and may well perform in small samples too. Thirdly, the ARDL approach generally yields unbiased long-run estimates even when some of the regressors in the model are endogenous. Besides, long- and short-run coefficients may be synchronously estimated in the framework of the ARDL approach. Considering these superiorities, the econometric model in Eq. 1 is expressed in an ARDL form as follows:
In Eq. 2, all variables are as defined in Eq. 1, and ∆ symbolizes the first difference operator. According to the ARDL approach procedure, in the first stage, based on the joint F-statistic, bounds testing is performed for cointegration analysis. Because the asymptotic distribution of F-statistics is non-standard under the null hypothesis which refers to no cointegration between variables, Pesaran et al. (2001) composed two sets of critical values for a given significance level, i.e., lower and upper critical values. As to the decision for cointegration, if the computed test statistic exceeds the upper critical bounds value, the existence of cointegration is confirmed. On the contrary, there is no cointegration in the case of computed test statistic is lower than the lower critical bounds value. As a third alternative, computed statistics may fall into upper and lower bounds, and then a clear inference concerning the cointegration cannot be made. Thus, to pass to estimation of parameter coefficients, at first, the existence of cointegration must be affirmed. Ultimately, the study estimates the model in Eq. 2 as per the ARDL analysis procedure, and the results are exhibited in Table 3.
Table 3 Empirical Results Based on the ARDL Approach
|
Predictors |
Model (1) |
Model (2) |
Model (3) |
Model (4) |
Model (5) |
Model (6) |
|
ln(GDPpc) |
0.079*** [0.033] |
-- |
-- |
-- |
-- |
-- |
|
ln(IND) |
-- |
2.905* [0.489] |
-- |
2.354* [0.417] |
1.867* [0.455] |
-- |
|
ln(FDI) |
0.544** [0.102] |
-- |
0.253* [0.034] |
-- |
-- |
0.766* [0.056] |
|
ln(MHTE) |
-- |
-- |
-3.758* [0.733] |
-- |
-- |
-- |
|
ln(REC) |
-- |
-- |
-- |
0.149 [0.127] |
-- |
-- |
|
ln(CR) |
-- |
-0.081*** [0.037] |
-- |
-- |
--
|
-0.164* [0.015] |
|
ln(RD) |
-- |
-- |
-- |
-- |
-0.205 [0.155] |
-- |
|
Constant |
1.287* [0.139] |
-3.106* [0.807] |
8.217* [1.298] |
-2.378* [0.747] |
-1.356 [0.771] |
1.337* [0.026] |
|
Selected Model |
(4,4,3) |
(1,3,2) |
(2,4,2) |
(1,3,0) |
(1,0,3) |
(4,3,4) |
|
Ramsey Reset Test |
0.894 |
1.375 |
1.009 |
2.801 |
2.491 |
2.515 |
|
Bounds Test |
25.451* |
4.389* |
6.271* |
3.643*** |
3.743*** |
147.151* |
Note: *, **, and *** indicate statistical significance at 1%, 5%, and 10% levels, respectively. Standard errors are in [brackets].
As seen in Table 3, Eq. 2 is hierarchically estimated to primarily avoid the multicollinearity problem. When reviewing the correlation coefficients between independent variables in Table 1, it is seen that most variables are highly correlated with each other. This problem weakens the robustness of estimation and most likely leads to biased and inefficient results. Therefore, explanatory variables are separately included in the estimation by considering their correlation relationships. Another reason for hierarchical estimation is the lack of a sufficient number of observations, which restricts the inclusion of all variables in the estimation simultaneously. The study estimates six different model specifications determined based on correlation results. For all models, after ascertaining a suitable lag number, the ARDL models are selected according to the Akaike Information Criteria (AIC). Ramsey Reset test results refer to the non-existence of model specification mistakes for all models. Performing bounds tests for cointegration shows that all model specifications are significant. In other words, null hypotheses are rejected because the computed F-statistics are higher than upper critical bounds values. That means the existence of cointegration among examined variables for all model specifications and enables us to estimate the long-run coefficients.
Empirical findings of the six models are summarized as follows: (i) GDP per capita is statistically significant yet has a very low positive effect on PM2.5 concentrations. Being in line with those of Li et al. (2016), Han et al. (2016), and Ahmad et al. (2020), but not with those of Zhou et al. (2018), this finding may be interpreted as the Chinese economy did not all that grow in an environmentally friendly way and growth dynamics embody some polluting components, even so a weak spurring impact of economic growth on PM2.5 concentrations. Yet, it should be kept in mind that this result does not enable us to make inferences regarding the position of the Chinese economy in the EKC pattern. To detect the clear position of growth of the Chinese economy in the way of the EKC path, the square of economic growth should be incorporated into the model. This is out of the scope of this research due to principally methodological restrictions. (ii) In some way of supporting this finding, in recent times considerably contributing to the economic development path of China; FDI inflows are statistically significant and affect PM2.5 concentrations positively. Being compatible with those of Ren et al. (2014), Jiang (2015), and Liu et al. (2019), this proof is substantially robust given the fact that in all model specifications, FDI inflows turn out to be statistically significant and give rise to more PM2.5 concentrations, and confirms that the pollution heaven hypothesis is valid for China in the relevant period. (iii) The industrialization is statistically significant in all model specifications and has a strong positive influence on PM2.5 concentrations. This finding is aligned with those of Li et al. (2016), Zhou et al. (2018), and Liu et al. (2019). As a result, industrialization is evaluated as the primary source of the air pollution problem proxied by PM2.5 concentrations in China.
(iv) Medium- and high-tech exports are found significant and have a strong negative impact on PM2.5 concentrations. This finding indicates that medium- and high-technology exports were the primary drivers of the declining trend in PM2.5 concentrations over the relevant timeframe. As a result, it can be stated that one of the most important supporters of China's APPCAP policy is medium- and high-tech exports, which symbolize the technological transformation in China's foreign trade. This result may most likely shed light on the mystery of the declining trend in PM2.5 concentrations in China in recent years. On the other hand, (v) unexpectedly renewable energy consumption is not one of the underlying dynamics of PM2.5 concentrations in China because it is found statistically insignificant. (vi) A similar evaluation may be made for research and development expenditures that have no significant impact on PM2.5 concentrations. Recently, given the fact that a considerable amount of research and development expenditures in the energy domain have been allocated for renewable energy, these insignificant findings are consistent with each other. Lastly, (vii) estimation results refer to the statistically significantly disincentive role of coal rents on PM2.5 concentrations. At first sight, this may be considered an unexpected and surprising result because coal is among non-renewable energy types, which are known as the most polluting factors of air in the literature. However, the study does not take coal production or consumption into account in a traditional style; instead, coal rents are considered owing to particularly seeking a comprehensive answer to the question "Is the NRC hypothesis valid for China from an environmental perspective". Results indicate that China gets rid of the NRC hypothesis from an environmental perspective by utilizing the rents from coal sources in the adoption and extension of a less-polluting production and trade structure.
At the beginning of the 2000s, being a notable emerging market economy, China significantly made progress in several components of its economic development path, such as GDP per capita, FDI inflows, and trade volume. However, in parallel with these achievements, increasingly rising concerns about air pollution in China have called for taking some steps to policy-makers. In return for this calling, in 2013, the Chinese State Council put into action the APPCAP to promote manufacturing technology in environmentally friendly industries by rewarding firms with high-energy efficiency and low emissions, providing business income tax incentives to high-tech companies with environmental protection projects, and offering financial support to air pollution control projects to promote innovation. Overall, on the one hand, the APPCAP has succeeded in alleviating air pollution; on the other hand, brought about some transformations in economic structure. Of course, these transformations have been reducing factors of air pollution. At this point, on the one side, while this study accepts the positive role of APPCAP in reducing air pollution; on the other side, the question ‘What elements contributed to this achievement' has made a sensation. In this context, in the framework of the ARDL approach, this study analyzed the underlying factors of air pollution represented with PM2.5 concentrations in China for the period 2000-2020, of which most are likely highly associated with APPCAP.
Analysis results strongly shed light on the factors of declining PM2.5 concentrations. Being important components of economic development, GDP per capita, industrialization, and FDI inflows act as air-polluting factors. In other words, improvements in these factors also bring about an unclear environment. From this window, either declining or structural transformation in these factors would lead to a clearer environment. In this context, in recent times, a downward trend in PM2.5 concentrations may be attributed to a declining trend in industrialization and FDI inflows. However, since they are important components of the economic development process, a necessary structural transformation in these factors is required. In other words, instead of cutting down on these factors, the industrialization process and FDI inflows should be more environmentally friendly. In some way guiding this argument, medium- and high-tech product exports are strongly and negatively associated with PM2.5 concentrations. It means that both industrialization structures should be toward medium- and high-tech products and FDI types specialized in medium- and high-tech goods should be encouraged. These recommendations seem to be applied within APPCAP, which suggests a similar way. Therefore, this plan should be strictly continued and even gone further. At the point of resource, coal rents may help to create a more qualified environment due to their negative impact on PM2.5 concentrations.
Ahmad F, Draz MU, Ozturk I, Su L, Rauf A (2020) Looking for asymmetries and nonlinearities: the nexus between renewable energy and environmental degradation in the Northwestern provinces of China. Journal of Cleaner Production 266: 1-17. https://doi.org/10.1016/j.jclepro.2020.121714
Akça EE (2024) Do renewable energy sources improve air quality? Demand-and supply-side comparative evidence from industrialized and emerging industrial economies. Environmental Science and Pollution Research 31(1): 293-311. https://doi.org/10.1007/s11356-023-30946-2
Auty R (1993) Sustaining development in mineral economies: The resource curse thesis. Oxford University Press, New York.
Cheng Z, Li L, Liu J (2017) Identifying the spatial effects and driving factors of urban PM2.5 pollution in China. Ecological Indicators 82: 61-75. https://doi.org/10.1016/j.ecolind.2017.06.043
Cheng L, Zhang T, Chen L, et al (2020) Investigating the impacts of urbanization on PM2.5 pollution in the Yangtze River Delta of China: A spatial panel data approach. Atmosphere 11(10): 1-17. https://doi.org/10.3390/atmos11101058
Cheng Y, Du K, Yao X (2023) Stringent environmental regulation and inconsistent green innovation behavior: Evidence from air pollution prevention and control action plan in China. Energy Economics 120: 1-12. https://doi.org/10.1016/j.eneco.2023.106571
Cole MA (2004) Trade, the pollution haven hypothesis, and the environmental Kuznets curve: examining the linkages. Ecological Economics 48(1): 71-81. https://doi.org/10.1016/j.ecolecon.2003.09.007
Ding D, Xing J, Wang S, Liu K, Hao J (2019) Estimated contributions of emissions controls, meteorological factors, population growth, and changes in baseline mortality to reductions in ambient PM 2.5 and PM2.5-related mortality in China, 2013–2017. Environmental Health Perspectives 127(6): 1-12. https://doi.org/10.1289/EHP4157
Feng Y, Ning M, Lei Y, et al (2019) Defending blue sky in China: Effectiveness of the “Air Pollution Prevention and Control Action Plan” on air quality improvements from 2013 to 2017. Journal of Environmental Management 252: 1-13. https://doi.org/10.1016/j.jenvman.2019.109603
Grimes P, Kentor J (2003) Exporting the greenhouse: Foreign capital penetration and CO2 emissions 1980 1996. Journal of World-Systems Research 9(2): 261–275. https://doi.org/10.5195/jwsr.2003.244
Grossman GM, Krueger AB (1996) The inverted-U: what does it mean? Environment and Development Economics 1(1): 119-122. https://doi.org/10.1017/S1355770X00000450
Guo J, Zhou Y, Ali S, Shahzad U, Cui L (2021) Exploring the role of green innovation and investment in energy for environmental quality: An empirical appraisal from provincial data of China. Journal of Environmental Management 292: 1-8. https://doi.org/10.1016/j.jenvman.2021.112779
Halkos GE, Tzeremes NG (2011) Growth and environmental pollution: Empirical evidence from China. Journal of Chinese Economic and Foreign Trade Studies 4(3): 144-157. https://doi.org/10.1108/17544401111178195
Han L, Zhou W, Li W (2016) Fine particulate (PM2.5) dynamics during rapid urbanization in Beijing, 1973–2013. Scientific Reports 6(1): 1-5. https://doi.org/10.1038/srep23604
Hao Y, Xu Y, Zhang J, et al (2019) Relationship between forest resources and economic growth: Empirical evidence from China. Journal of Cleaner Production 214: 848-859. https://doi.org/10.1016/j.jclepro.2018.12.314
Hoffmann R, Lee CG, Ramasamy B, Yeung M (2005) FDI and pollution: a granger causality test using panel data. Journal of International Development: The Journal of the Development Studies Association 17(3): 311-317. https://doi.org/10.1002/jid.1196
Hsu CC, Zhang Y, Ch P, et al (2021) A step towards sustainable environment in China: The role of eco-innovation renewable energy and environmental taxes. Journal of Environmental Management 299: 1-10. https://doi.org/10.1016/j.jenvman.2021.113609
Huang J, Pan X, Guo X, Li G (2018) Health impact of China's Air Pollution Prevention and Control Action Plan: An analysis of national air quality monitoring and mortality data. The Lancet Planetary Health 2(7): 313-323. https://doi.org/10.1016/S2542-5196(18)30141-4
International Energy Agency (2023) World energy investment report. https://www.iea.org/reports/world-energy-investment-2023 Accessed 26 December 2023
IQAir (2022) World air quality report. https://www.iqair.com/world-air-quality-report Accessed 13 July 2023
Jiang Y (2015) Foreign direct investment, pollution, and the environmental quality: a model with empirical evidence from the Chinese regions. The International Trade Journal 29(3): 212-227. https://doi.org/10.1080/08853908.2014.1001538
Jiang P, Yang J, Jiang P, et al (2018) The contribution of socioeconomic factors to PM2.5 pollution in urban China. Environmental Pollution 233: 977-985. https://doi.org/10.1016/j.envpol.2017.09.090
Jorgenson AK (2007) Does foreign investment harm the air we breathe and the water we drink? A cross-national study of carbon dioxide emissions and organic water pollution in less-developed countries, 1975 to 2000. Organization & Environment 20(2): 137-156. https://doi.org/10.1177/1086026607302153
Li G, Fang C, Wang S, Sun S (2016). The effect of economic growth, urbanization, and industrialization on fine particulate matter (PM2.5) concentrations in China. Environmental Science & Technology 50(21): 11452-59. https://doi.org/10.1021/acs.est.6b02562
Lin B, Zhu J (2019) The role of renewable energy technological innovation on climate change: Empirical evidence from China. Science of the Total Environment 659: 1505-12. https://doi.org/10.1016/j.scitotenv.2018.12.449
Liu Q, Wang S, Zhang W, Li J, Dong G (2019) The effect of natural and anthropogenic factors on PM2.5: Empirical evidence from Chinese cities with different income levels. Science of the Total Environment 653: 157-167. https://doi.org/10.1016/j.scitotenv.2018.10.367
Odhiambo NM (2009) Energy consumption and economic growth nexus in Tanzania: An ARDL bounds testing approach. Energy Policy 37(2): 617-622. https://doi.org/10.1016/j.enpol.2008.09.077
OECD (2023) Exposure to PM2.5 in countries and regions. https://stats.oecd.org/Index.aspx?DataSetCode=EXP_PM2_5 Accessed 26 November 2023
Perkins R, Neumayer E (2008) Fostering environment efficiency through transnational linkages? Trajectories of CO2 and SO2, 1980–2000. Environment and Planning A 40(12): 2970-89. https://doi.org/10.1068/a4089
Pesaran MH, Shin Y, Smith RJ (2001) Bounds testing approaches to the analysis of level relationships. Journal of Applied Econometrics 16(3): 289-326. https://doi.org/10.1002/jae.616
Ren S, Yuan B, Ma X, Chen, X (2014) International trade, FDI (foreign direct investment) and embodied CO2 emissions: A case study of China’s industrial sectors. China Economic Review 28: 123-134. https://doi.org/10.1016/j.chieco.2014.01.003
Shen G, Ru M, Du W, et al (2019) Impacts of air pollutants from rural Chinese households under the rapid residential energy transition. Nature Communications 10(3405): 1-8. https://doi.org/10.1038/s41467-019-11453-w
Shen Y, Su ZW, Malik MY, et al (2021) Does green investment, financial development and natural resources rent limit carbon emissions? A provincial panel analysis of China. Science of the Total Environment 755: 1-12. https://doi.org/10.1016/j.scitotenv.2020.142538
Talukdar D, Meisner CM (2001) Does the private sector help or hurt the environment? Evidence from carbon dioxide pollution in developing countries. World Development 29(5): 827-840. https://doi.org/10.1016/S0305-750X(01)00008-0
UNEP (2023) Pollution action note – Data you need to know. https://www.unep.org/interactive/air-pollution-note/ Accessed 26 December 2023
Wang DT, Chen WY (2014) Foreign direct investment, institutional development, and environmental externalities: Evidence from China. Journal of Environmental Management 135: 81-90. https://doi.org/10.1016/j.jenvman.2014.01.013
Wang H, Zhang Y, Zhao H, et al (2017) Trade-driven relocation of air pollution and health impacts in China. Nature Communications 8(738): 1-7. https://doi.org/10.1038/s41467-017-00918-5
Wang Y, Bi F, Zhang Z, et al (2018) Spatial production fragmentation and PM2.5 related emissions transfer through three different trade patterns within China. Journal of Cleaner Production 195: 703-720. https://doi.org/10.1016/j.jclepro.2018.05.195
Wang Y, Cao J (2023) Examining the effects of socioeconomic development on fine particulate matter (PM2.5) in China’s cities based on spatial autocorrelation analysis and MGWR model. International Journal of Environmental Research and Public Health 20(4): 1-23. https://doi.org/10.3390/ijerph20042814
Wheeler D (2001) Racing to the bottom? Foreign investment and air pollution in developing countries. The Journal of Environment & Development 10(3): 225-245. https://doi.org/10.1177/10704965-0101003-02
World Bank (2023) World Development Indicators. https://databank.worldbank.org/source/world-development-indicators Accessed 26 December 2023
Xu B, Luo L, Lin B (2016). A dynamic analysis of air pollution emissions in China: Evidence from nonparametric additive regression models. Ecological Indicators 63: 346-358. https://doi.org/10.1016/j.ecolind.2015.11.012
Xu SC, Miao YM, Gao C, et al (2019) Regional differences in impacts of economic growth and urbanization on air pollutants in China based on provincial panel estimation. Journal of Cleaner Production 208: 340-352. https://doi.org/10.1016/j.jclepro.2018.10.114
Xu N, Zhang F, Xuan X (2021) Impacts of industrial restructuring and technological progress on PM2.5 pollution: evidence from prefecture-level cities in China. International Journal of Environmental Research and Public Health 18(10): 5283. https://doi.org/10.3390/ijerph18105283
Yan D, Ren X, Zhang W, et al (2022) Exploring the real contribution of socioeconomic variation to urban PM2.5 pollution: New evidence from spatial heteroscedasticity. Science of the Total Environment 806: 1-11. https://doi.org/10.1016/j.scitotenv.2021.150929
Zeduo Z, Wei Z, Mou Y, et al (2022) Evolutionary relationship between the PM2. 5 concentration and urbanization in the Yellow River Ecological and Economic Zone. Environmental and Sustainability Indicators 16: 1-9. https://doi.org/10.1016/j.indic.2022.100198
Zhang C, Zhou X (2016) Does foreign direct investment lead to lower CO2 emissions? Evidence from a regional analysis in China. Renewable and Sustainable Energy Reviews 58: 943-951. https://doi.org/10.1016/j.rser.2015.12.226
Zhang X, Xu X, Ding Y, et al (2019) The impact of meteorological changes from 2013 to 2017 on PM 2.5 mass reduction in key regions in China. Science China Earth Sciences 62: 1885-1902. https://doi.org/10.1007/s11430-019-9343-3
Zhao B, Zheng H, Wang S, et al. (2018) Change in household fuels dominates the decrease in PM2.5 exposure and premature mortality in China in 2005–2015. Proceedings of the National Academy of Sciences 115(49): 12401-12406. https://doi.org/10.1073/pnas.1812955115
Zhou C, Chen J, Wang S (2018) Examining the effects of socioeconomic development on fine particulate matter (PM2.5) in China's cities using spatial regression and the geographical detector technique. Science of the Total Environment 619: 436-445. https://doi.org/10.1016/j.scitotenv.2017.11.124
Zhou L, Tang L (2021) Environmental regulation and the growth of the total-factor carbon productivity of China's industries: Evidence from the implementation of action plan of air pollution prevention and control. Journal of Environmental Management 296: 1-12. https://doi.org/10.1016/j.jenvman.2021.113078
Zhu X, Yun X, Meng W, et al (2019) Stacked use and transition trends of rural household energy in mainland China. Environmental Science & Technology 53(1): 521-529. https://doi.org/10.1021/acs.est.8b04280
The authors declare no competing interests.