Research PaperFree Access

Explaining Provincial Variation in Implementation of China’s Clean Water Policies

Fang Xu

Department of Politics and International Relations, The University of Auckland, 58 Symonds Street, Auckland 1010, New Zealand

E-mail Address: fxu332@aucklanduni.ac.nz

Search for more papers by this author

https://doi.org/10.1142/S2382624X22500096Cited by:4 (Source: Crossref)

Abstract

This research analyzes the local implementation of China’s central clean water policies over the recent decades. Based on a series of panel data analyses on Chinese provinces between 2004 and 2015, this study empirically examines the impact of decentralization and interest groups on water policy implementation, leading to three main findings. First, fiscal decentralization has a significant positive effect on policy output as measured using per-capita provincial expenditures on industrial wastewater. But it has no significant influence on policy outcome as measured by per-capita emissions of chemical oxygen demand. Second, while increasing environmental decentralization at the provincial level tends to increase provincial expenditures, wastewater discharge also increases. Third, increased industrial contributions to the provincial economy are associated with increased environmental spending from the province, yet foreign trade and environmental petitions do not have the expected impact. The results suggest that China’s decentralization reforms appear to have increased provincial policy responses without improving the actual environmental outcomes. Experiences from authoritarian China may provide lessons for other countries.

Keywords:

1. Introduction

Using panel data for Chinese provinces between 2004 and 2015, this study focuses on the impact of fiscal decentralization, environmental decentralization, and interest group pressure to help account for provincial variation in the implementation of China’s clean water policies. Water source protection and economic growth are often considered competing objectives in China and other developing countries (Hu and Cheng 2013; Joshi and Beck 2015). A direct relationship between water pollution and local economic growth in China has been established in recent studies. For example, Wang et al. (2006) argue that the increasing discharge of chemical oxygen demand (COD; a common tool to assess water pollution) in the Taihu Basin is closely related to rapid urbanization and industrialization there. However, this pattern may vary depending upon the overall development and wealth within regions. Li et al. (2016) observe that the intensity of water pollution emissions declined as the economy grew in China’s coastal areas during 2003–2010. This is because more-developed regions face more pressure related to long-term sustainability compared to less-developed regions.

Factors that shape and influence local water policy implementation are multifaceted and vary in place and time (Han et al.2016; Wang et al.2008). In China, the implementation of environmental policies is closely related to its fiscal and environmental decentralization reforms (Chen and Lees 2018; Qi et al.2014).

Fiscal decentralization refers to the handover of public fiscal rights and government services from the central government to local governments (Ma and Mao 2018; Yong 2010). Fiscal decentralization in China is focused on the autonomy granted by the central government to local governments in terms of tax revenue, economic development, and resource allocation (Lin and Liu 2000). This provides incentives for local governments to pursue market-supporting activities (Ma and Mao 2018). However, decentralization within China leads the central government to use local economic growth to evaluate the performance of local cadres (You et al.2019). Under certain circumstances, this may lead to a “race to the bottom” between local authorities as cadres are more likely to promote economic growth at the expense of the environment (Cai et al.2016).

Recent studies further show that decentralization is a complex process involving not only financial but also governance aspects (Azfar et al.2018; Qi et al.2014). China has implemented reforms to improve the environmental protection management system owing to pollution problems caused by population growth and economic development (Wu et al.2020). The environmental decentralization reforms aim to reasonably distribute the power and responsibility for environmental protection to governments at all levels (Sigman 2007). These levels are rooted in China’s top–down administrative system which is characterized by the participation of different agencies and officials geographically dispersed within various levels of the government involved in environmental policy implementation (Han et al.2016). They have different interests, preferences, goals, and tools, which may variably influence their environmental performance (Qi and Zhang 2014).

China’s decentralized governance model demonstrates that local authorities have fiscal and administrative authority to influence local environmental decision-making (Chen and Lees 2018; Qi et al.2014). When fiscal decentralization and environmental decentralization follow an internally consistent logic of decentralization as a whole, they inevitably affect local environmental strategies. An increase in local governments’ autonomy in fiscal or environmental governance tends to reduce the environmental regulation standards (Mol and Carter 2006) and aggravate environmental pollution (Zhang et al.2017). The main reason for these phenomena is to improve the local fiscal revenue.

Although some studies, both outside of China (Farzanegan and Mennel 2012) and within (Lan et al.2014), argue that decentralization has negative regulatory effects on environmental management, other authors firmly disagree. Based on the data of 47 countries from 1979 to 1999, Sigman (2007) finds that fiscal decentralization did not lead to environmental quality deterioration. Millimet (2003) indicates that environmental decentralization has promoted better pollution control policies and mitigated the deterioration of air pollution in the United States. Several other scholars report that fiscal decentralization has no significant effect on environmental pollution (Hao et al.2020). Using Chinese provincial panel data for 1995–2010, He (2015) argues that while fiscal decentralization has no significant effect on wastewater emissions, it has a significant and positive effect on pollution abatement spending and pollutant discharge fees. These conflicting views and evidence suggest the relative importance of fiscal decentralization and environmental decentralization and need further evaluation.

Contrary to formal participants in policy processes, attention to local interest groups, including industries, environmental organizations, and the public, has grown (Li et al.2012; Wang et al.2008; Xie 2016). Recently, scholars have studied the role of interest groups in environmental governance, focusing mainly on the following perspectives. The first is the negative impacts resulting from government–enterprise collusion (Lu 2021; Zhang et al.2017). To attract investments and promote local economic growth, provincial governments made exceptions for high-energy-dissipation and high-emission enterprises by weakening the environmental policies from 1995 to 2012 (Zhang et al.2017). Although local Chinese governments partially subsidize environmental expenditures of industrial polluters to achieve cleaner production, pollution emissions have not been reduced (Dong et al.2010). Scholars suggest that the more number of industrial enterprises, the higher the possibility of water pollution (Wang et al.2008). Furthermore, several authors argue that international trade tends to shift environmental pollution to developing countries like China (Perkins and Neumayer 2009). Jiang and Zheng (2017) find that foreign trade is positively related to pollution intensity in the cities of Yangtze River Delta. Ljungwall and Linde-Rahr (2005) provide an example wherein they argue that less-developed regions in China prefer policies for attracting foreign direct investment than those for the environment.

In addition to the influence of local and foreign industries, with the growing environmental crisis, public involvement in environmental regulation has increased significantly over the last decade (Li et al.2012). In China, the public can participate via party members, complain to mass media or government offices, and may protest openly under extreme circumstances (Wu et al.2018). Although recent studies suggest that environmental groups may effectively oppose polluting industries and push for stricter local enforcement (Bernauer et al.2016; Xie 2016), authors do not necessarily agree on the public’s role in China’s environmental management (Kostka and Mol 2013; Yew and Zhu 2019). One prominent explanation shows that the central government uses top–down environmental management tools and creates limited political space for public participation (Yew and Zhu 2019). Given the different views expressed, the impact of factors representing aspects of local interest groups needs to be further assessed.

This study explores how fiscal and environmental decentralization levels and pressure groups influence water policy implementation at the provincial level. This research addresses a crucial empirical issue for post-reform China and an important theoretical question relevant to all countries that have to strike a balance between centralization and decentralization regarding water governance. This paper is organized into four sections. Section 2 presents the primary data sources, models, variables, and methods. Section 3 reports the results of the study. Section 4 concludes with a discussion of their implications.

2. Materials and Methods

2.1. Data sources

Data used in the following analysis are obtained from several sources published in China from 2004 to 2015. These include the China Statistical Yearbooks on Environment, Statistical Yearbooks of China, and China Industry Statistical Yearbooks. Data on local per-capita expenditures, pollutants, environmental impact assessments, environmental management personnel, environmental petitions, and water-use intensity come from the China Statistical Yearbooks on Environment. Data about provincial budgetary expenditures and per-capita GDP come from the Statistical Yearbooks of China. Statistics on various industries and regions are from the China Industry Statistical Yearbooks. Data of provincial leaders come from local official websites.

2.2. Model specification

Following existing studies with panel data analyses (Landry et al.2018; Zhang 2017), this study adopts a fixed-effect approach to control for region-specific and time-specific effects.¹ The fixed-effect model includes the following advantage. First, each province is a separate cross-sectional entity, and economic and social development characteristics may be significantly different among all these entities. The fixed-effect method can adjust for differences among provinces. Second, while the number of cross-sectional individuals in panel data is limited, the total sample is not large. The fixed-effect method avoids the loss of degrees of freedom. This study uses robust linear mixed models (R package robustlmm) to address the possible impact of heteroscedasticity.² We use the following formula to conduct the panel data analyses (the specific variables used in this formula will be defined in the following subsection) :

y i t = α Finance i t + β EIA i t + γ EM i t + δ Industry i t + ϵ Foreign i t + ε Petition i t + ζ X i t + μ i + φ t + ν i t . \begin{matrix} y_{i t} & = & α {Finance}_{i t} + β {EIA}_{i t} + γ {EM}_{i t} + δ {Industry}_{i t} + ϵ {Foreign}_{i t} + ε {Petition}_{i t} + ζ X_{i t} + μ_{i} + φ_{t} + ν_{i t} . \end{matrix} <math display="block" altimg="eq-00001.gif"><mrow><mtable displaystyle="true" equalrows="false" equalcolumns="false"><mtr><mtd columnalign="right"><msub><mrow><mi>y</mi></mrow><mrow><mi>i</mi><mi>t</mi></mrow></msub></mtd><mtd columnalign="center"><mo>=</mo></mtd><mtd columnalign="left"><mi>α</mi><msub><mrow><mstyle><mtext mathvariant="normal">Finance</mtext></mstyle></mrow><mrow><mi>i</mi><mi>t</mi></mrow></msub><mo>+</mo><mi>β</mi><msub><mrow><mstyle><mtext mathvariant="normal">EIA</mtext></mstyle></mrow><mrow><mi>i</mi><mi>t</mi></mrow></msub><mo>+</mo><mi>γ</mi><msub><mrow><mstyle><mtext mathvariant="normal">EM</mtext></mstyle></mrow><mrow><mi>i</mi><mi>t</mi></mrow></msub><mo>+</mo><mi>δ</mi><msub><mrow><mstyle><mtext mathvariant="normal">Industry</mtext></mstyle></mrow><mrow><mi>i</mi><mi>t</mi></mrow></msub><mo>+</mo><mi>ϵ</mi><msub><mrow><mstyle><mtext mathvariant="normal">Foreign</mtext></mstyle></mrow><mrow><mi>i</mi><mi>t</mi></mrow></msub></mtd></mtr><mtr><mtd columnalign="right"></mtd><mtd columnalign="center"></mtd><mtd columnalign="left"><mo>+</mo><mspace width=".17em"></mspace><mi>ε</mi><msub><mrow><mstyle><mtext mathvariant="normal">Petition</mtext></mstyle></mrow><mrow><mi>i</mi><mi>t</mi></mrow></msub><mo>+</mo><mi>ζ</mi><msub><mrow><mi>X</mi></mrow><mrow><mi>i</mi><mi>t</mi></mrow></msub><mo>+</mo><msub><mrow><mi>μ</mi></mrow><mrow><mi>i</mi></mrow></msub><mo>+</mo><msub><mrow><mi>φ</mi></mrow><mrow><mi>t</mi></mrow></msub><mo>+</mo><msub><mrow><mi>ν</mi></mrow><mrow><mi>i</mi><mi>t</mi></mrow></msub><mo>.</mo></mtd></mtr></mtable></mrow></math>

Here,

y_{i t}

$y_{i t}$ is the indicator for provincial expenditure or measures of water pollutants discharge in jurisdiction

i

$i$ at year

t

$t$ .

{Finance}_{i t}

${Finance}_{i t}$ is the indicator of fiscal decentralization.

{EIA}_{it}

${EIA}_{it}$ is the relative weight of provincial administrative capacity for environmental impact assessment (EIA).

{EM}_{it}

${EM}_{it}$ is the relative provincial power of environmental monitoring (EM).

{Industry}_{i t}

${Industry}_{i t}$ ,

{Foreign}_{i t}, and {Petition}_{i t}

${Foreign}_{i t}, and {Petition}_{i t}$ indicate the influences of local industries, foreign trade, and public concern, respectively.

X_{i t}

$X_{i t}$ is a set of control variables. Variable

μ_{i}

$μ_{i}$ is the province-specific effect, and

φ_{t}

$φ_{t}$ is the year-specific effect. The error term

ν_{i t}

$ν_{i t}$ captures the effects of other disturbances on the dependent variables. To meet the assumption of an approximately normal distribution of variables, the natural logarithm is used to transform highly skewed variables in the estimation. We have performed the preliminary examination of residuals, and the natural log transformation has the intended impact.

2.3. The variables and hypotheses

In this study, policy implementation is operationalized at two different levels: policy output and policy outcome. Policy implementation is essential for effective environmental governance (van Rooij et al.2017). As a complex phenomenon, the implementation may be best explained and understood as a process with both outputs and outcomes (Hill and Hupe 2002). These represent two main measures of implementation. Measuring policy implementation in terms of “outputs” refers to the assessment of actions and efforts of local governments. In contrast, “outcomes” refers to the results influenced by those actions.

Policy output is measured here by per-capita expenditures for industrial wastewater treatment in each province in the first level. Government expenditure on environmental issues is a commonly accepted operational concept for policy implementation, though recognized as imperfect (Stegarescu 2005). The amount of expenditure is an imperfect measurement here because of the opacity of China’s political process. Nevertheless, this indicator may still help demonstrate the willingness and preferences of localities to allocate resources towards the defined goals. We anticipate that greater local per-capita expenditures represent higher levels of provincial enforcement of clean water policies.

To address the second level, we treat the per-capita volume of industrial wastewater or per-capita COD discharge as indicators of policy outcomes. Several authors argue that a water governance’s relative success or failure is more effectively assessed using evidence of change in the water quality outcomes than measures of outputs (Biddle and Koontz 2014; Koontz and Thomas 2006). Since the 1970s, China has increased its focus and efforts towards controlling industrial wastewater pollution (Han et al.2016). COD has been regarded as one of the most important measures of water pollution in national plans for decades (Kahn et al.2015). Thus, examining both expenditures and COD indicators allows this study to identify and evaluate uneven implementation among the provinces.

We use expenditure decentralization to represent fiscal decentralization. This is defined as the ratio of provincial budgetary expenditures to central government fiscal expenditures. Existing literature on the measurement of fiscal decentralization can be divided into three categories related to the operational variable used: fiscal revenue, fiscal expenditure, and a fiscal autonomy index (Stegarescu 2005). Among them, an expenditure indicator is often used as a proxy for the transfer of fiscal powers and responsibilities from the center to localities (He 2015). Studies suggest that increasing the fiscal autonomy of local authorities will make the implementation of environmental policies more difficult as local officials tend to sacrifice environmental regulations to boost economic growth (Cai et al.2016; Zhang 2017; Zhang and Fu 2008). Thus, we hypothesize the following.

$H_{1}$ . Higher levels of fiscal decentralization within provinces lead to the reduced implementation of clean water policies as assessed by the provincial per-capita expenditures on wastewater treatment or two measures of water pollution.

Most studies examine environmental policies and their environmental impact from a “fiscal decentralization” perspective (Hao et al.2020; Kuai et al.2019). However, fiscal decentralization cannot replace environmental decentralization owing to the unique characteristics of environmental protection affairs (Chen and Lees 2018). In this regard, this study adopts the idea of some more recent studies by Ma (2019) and Wu et al. (2020) and uses changes in the adjusted ratio of EIA cases approved at the provincial versus central levels, as well as the adjusted ratio of provincial to central government EM personnel to characterize the dynamic changes in environmental decentralization in the Chinese context.³

The EIA has been a key component of China’s formal decision-making process in all types of governance (Ma 2019; Zhou and Sheate 2011). The decentralization of EIA demonstrates the relative weight of environmental decision-making power between central government and local governments. On the one hand, the decentralization process has the potential to reduce the burden on the central government and accelerate the approval process (Ma 2019). On the other hand, case studies in China suggest that EIA power may be misused if local governments intend to achieve economic efficiency even at the expense of environmental pollution (Beyer 2006). Therefore, relatively more EIA cases approved at the provincial level may lead to lax implementation of clean water policies.

Provincial governments have prepared the necessary environmental monitoring agencies and train personnel in enforcing the environmental policies (Song et al.2010). Therefore, monitoring agencies and personnel to some extent reflect the government’s ability to exercise effective environmental supervision (Wu et al.2020). However, under the decentralization of environmental monitoring, when power is concentrated at the local level, provincial governments may become lenient towards polluting businesses. This results in ineffective enforcement (van Rooij et al.2017). Therefore, we present the following hypotheses.

$H_{2 a}$ . Increasing decentralization of EIA decreases the implementation of clean water policies as judged using measures of provincial per-capita expenditures or two measures of water pollution.

$H_{2 b}$ . More personnel monitoring environmental outcomes at the provincial level versus the central level are also anticipated to decrease the implementation of clean water policies using either provincial per-capita expenditures or two measures of pollutant discharge.

**Table 1. Summary of Variables for Models with Provincial Expenditure and Water Pollution as the Dependent Variables**
	Dependent Variables	Explanatory Variables	Control Variables
Policy output	Provincial industrial wastewater treatment expenditure(expenditure for industrial wastewater treatment in each province per capita)	Local expenditure decentralization(the ratio ofprovincial budgetary expenditure to central government budgetary expenditure)	Lag industrial COD discharge(one-year-lag industrial COD discharges per capita)Lag industrial wastewater discharge(one-year-lag industrial wastewater discharges per capita)
		Environmental impact assessment decentralization(cases approved at the provincial vs. central levels)	Economic development(per-capita GDP)
		Environmental monitoring decentralization(the ratio of provincial to central government environmental law enforcement personnel)	Lag government expenditure(one-year-lag provincial industrial wastewater treatment expenditure)
		Industrial strength(industry’s share of total GDP)	Provincial leaders(Governor & Party Secretary)[Age, Tenure, $i . e .$ , length of current Position, andEducation(0: high school; 1: college-level education; 2: post-graduate degree;and 3:doctoral degree)]
		Lag environmental petition(one-year-lag submission of environmental petitions per capita)
		Foreign trade(total export–import volume/GDP)
Policy outcomes	Industrial wastewater discharge(industrial wastewater discharges per capita)	Environmental impact assessment decentralization(cases approved at the provincial vs. central levels)	Provincial industrial wastewater treatment expenditures(expenditure for industrial wastewater treatment in each province per capita)
	Industrial COD discharge(industrial COD discharges per capita)	Environmental monitoring decentralization(the ratio of provincial to central government environmental law enforcement personnel)	Economic development(per-capita GDP)
		Industrial strength(industry’s share of total GDP)	Water-use intensity(volumes of industrial water per unit of industrial value-added)
		Environmental petition(one-year-lag submission of environmental petitions per capita)	Provincial leaders(Governor & Party Secretary)[Age, Tenure, $i . e .$ , length of current position, andEducation(0: high school; 1: college-level education; 2: post-graduate degree; and 3: doctoral degree)]
		Foreign trade (total export–import volume/GDP)

While the influence of interest groups on environmental issues might be less profound in China than that in Western countries, these groups are playing increasingly important roles (Li et al.2012; Wang et al.2008). We use three indicators to measure interest groups’ influences.

The first one is the influence of the polluting industry, which is measured by the share of industrial value-added in provincial GDP (Hao et al.2016; Tschofen et al.2019). Environmental regulation may cost the local industries a significant loss of profit if the industries are required to contribute funds (Wang et al.2008). A study in Italy shows that local industries exert pressure on governments for greater governmental environmental expenditures (Fiorino and Ricciuti 2009). This occurs owing to local governments aiming to create and maintain local industrial employment by providing environmental spending. Moreover, this may also apply to Chinese provinces. Meanwhile, industries with a large share of the GDP in the total output of a province, and thus influence, are likely to produce more water pollution (Zhang et al.2012).

The second indicator is foreign trade, which is measured by the share of total export–import volume as part of provincial GDP. This measurement is consistent with those of other studies (Sheng 2007; Zhang et al.2018). A study in China’s major cities suggests that foreign trade contributes the most to air pollution because local government officials are likely to reduce the environmental stringency of enterprises to decrease their cost of export products. This is believed to attract foreign enterprises and strengthen local market competitiveness in global trade (He et al.2012). We assume this may apply to water pollution in China as well.

The third indicator — the intensity of behaviors associated with public participation — is represented by the submission of environmental petitions per capita. A number of studies have demonstrated that bottom–up public participation puts pressure on the government for facing environmental conflicts (Li et al.2012; Xie 2016). The petitioning system is the only legitimate institution for Chinese people to express their complaints and concerns (Li 2008). Therefore, we explicitly assume that the greater the relative number of environmental petitions, the higher the degree of public participation in local environmental management. We use a one-year lag for this indicator owing to potential delay in their influences on expenditures and pollutants.

We propose the three following hypotheses.

$H_{3 a}$ . An increase in industrial contributions to provincial economies is associated with increased provincial per-capita expenditures on industrial wastewater or pollutant discharge assessed two ways.

$H_{3 b}$ . Greater foreign trade within provinces leads to fewer expenditures per capita and more water pollution as measured using volumes of COD and wastewater.

$H_{3 c}$ . More environmental petitions per capita in the preceding year are anticipated to lead to more provincial government expenditures per capita and reduced water pollutant emissions assessed two ways.

Given a dependent output variable (provincial per-capita expenditures on industrial wastewater) and two outcome variables (per-capita volumes of COD or wastewater), we conducted two sets of panel data analyses. In each set of regressions, there are two models: Models 1 and 2 for output and Models 3 and 4 for outcomes. In all models, we partially control for the level of economic development in provinces. Furthermore, some scholars have suggested that the personal characteristics of local politicians, including their age, tenure in their current position, and education level, may influence how environmental policies are implemented (Kostka and Yu 2015; Landry et al.2018); hence, we also include these variables as control variables.⁴

Models 1 and 2 include one-year-lag measures of water pollution and provincial expenditures as control variables. Scholars suggest that a more severe problem should make implementation more probable (Dombrowsky et al.2014). Thus, in data analysis, we include a one-year-lag industrial COD discharges per capita or one-year-lag industrial wastewater discharges per capita. This assumes that the severity of the water pollution problem in the past may affect local expenditures in the current year. Moreover, we also include a lagged dependent variable as local environmental expenditure levels are strongly influenced by the budgets set in the previous year (Duan and Zhan 2011; Zhu et al.2014).

Models 3 and 4 use the water-use intensity in terms of volumes of industrial wastewater per unit of industrial value-added to measure technological availability as a control variable. The effective implementation of water policies as measured by industrial wastewater discharge partially depends on the availability of existing technologies to control and reduce pollutants (Wang et al.2015). This indicator is used to measure the efficiency of water use in the industrial sector (Alcamo et al.2003). A decrease in the value of this indicator indicates improvements in technological efficiency. Improving water efficiency means increasing water productivity, i.e., reducing the intensity of water use and pollution from industrial activities (Flörke et al.2013). In addition, we use the expenditure for industrial wastewater treatment in each province per capita to measure how much importance a local government attaches to water policy enforcement. The provincial budget plan is a political message, which consists of information, expectations, and resources that influence the actions of different departments (Yong 2010). Hence, we assume that more government expenditures may reduce water pollution when a local government attaches greater attention to the enforcement of clean water policies. Table 2 presents the summary statistics from the final complete dataset used.

**Table 2. Summary Statistics of Provincial Variables**
Variables	Model(s)	Min	Median	Max	Mean	StandardDeviation
Dependent Variables (all natural-log-transformed)
Per-capita expenditure for industrial wastewater treatment	1 and 2	−1	0.94	1.75	0.93	0.38
Industrial COD discharges per capita	3	−0.69	1.04	2.31	1.01	0.49
Industrial wastewater discharges per capita	4	−0.41	0.72	1.56	0.73	0.33
Explanatory Variables
Decentralization
Local expenditure decentralization	1–4	0.02	0.16	0.50	0.17	0.08
Environmental impact assessment decentralization	1–4	0.53	19.18	359.01	36.62	43.97
Environmental monitoring decentralization	1–4	0.01	0.32	1.72	0.45	0.34
Interest groups
Industry’s share of total GDP	1–4	0.13	0.42	0.53	0.40	0.08
One-year-lag environmental petition	1–4	0	0.64	4.27	0.75	0.55
Foreign trade	1–4	0	0.01	0.15	0.02	0.04
Control Variables
Per-capita GDP (log)	1–4	3.64	4.45	5.03	4.43	0.29
One-year-lag provincial industrial wastewater treatment expenditure (log)	1 and 2	−1	0.93	1.75	0.93	0.38
One-year-lag COD (log)	1	−0.54	1.12	2.35	1.08	0.49
One-year-lag wastewater (log)	2	−0.37	0.78	1.63	0.79	0.33
Water-use efficiency (log)	3 and 4	−2.45	−1.77	−0.50	−1.77	0.30
Age_governor	1–4	46	59	65	58.22	3.83
Tenure length of current position_governor	1–4	1	3	10	3.13	1.87
Education_governor	1–4	0	2	3	1.7	0.79
Age $_{Party Secretary}$	1–4	47	61	70	59.90	4.26
Tenure length of current position $_{Party Secretary}$	1–4	1	3	15	3.24	2.31
Education $_{Party Secretary}$	1–4	0	2	3	1.46	0.92

3. Results

3.1. The impact of fiscal decentralization

The results support H₁, wherein higher levels of fiscal decentralization within provincial authorities lead to reduced water policy implementation output, as assessed by the per-capita industrial wastewater treatment expenditures. The regression coefficients and their standard errors reported in Table 3 suggest that the level of fiscal decentralization has a negative and statistically significant association with provincial per-capita wastewater expenditures. Provincial governments with more financial autonomy under a decentralized system tend to spend fewer public funds on industrial water pollution control. Table 4 shows that, unlike output, the initial hypothesized relationships between local expenditure decentralization and outcome measures are not supported. In Model 3, no apparent relationship is established between the local expenditure decentralization and per-capita COD discharge. In Model 4, provincial per-capita wastewater discharge volumes decrease with increasing levels of provincial expenditure decentralization, opposite to initial expectations ( $p = 0.050$ ).

**Table 3. Determinants of Provincial Per-Capita Industrial Wastewater Expenditure (2004–2015)**
	Model 1		Model 2
	Estimate (SE)	$p$ -Value	Estimate (SE)	$p$ -Value
Explanatory Variables
Decentralization
Local expenditure decentralization (H₁)	−0.582	0.046	−0.704	0.024
	(0.292)		(0.311)
Environmental impact assessment decentralization (H $_{2 a}$ )	0.001	0.021^a	0.001	0.033^a
	(0)		(0.001)
Environmental monitoringdecentralization (H $_{2 b}$ )	0.116	0.027^a	0.115	0.039^a
	(0.052)		(0.056)
Interest groups
Industrial strength (H $_{3 a}$ )	0.807	0.001	0.931	0.0007
	(0.249)		(0.274)
Foreign trade (H $_{3 b}$ )	0.091	0.890	−0.848	0.260
	(0.681)		(0.745)
One-year-lag environmental petition (H $_{3 c}$ )	−0.047	0.056	−0.044	0.089
	(0.025)		(0.026)
Control Variables
One-year-lag COD discharges per capita (log)	0.311	<0.0005
	(0)
One-year-lag wastewater discharges per capita (log)			0.347	<0.0005
			(0.098)
Per-capita GDP (log)	0.326	0.002	0.266	0.031
	(0.105)		(0.123)
One-year-lag provincial per-capita industrial wastewater treatment expenditure (log)	0.409	<0.0005	0.418	<0.0005
	(0.041)		(0.042)
Governor characteristics^b	Yes^c		Yes
Party Secretary characteristics^b	Yes		Yes
Province fixed effects	Yes		Yes
Year fixed effects	Yes		Yes
Numbers of observations	360		360

Notes: ^aStatistical significance is being reported here as if the hypotheses are two-sided. The outcomes of H $_{2 a}$ and H $_{2 b}$ are in the opposite direction of those originally anticipated. ^bThe variables to measure Governor and Party Secretary characteristics are age, education level, and the length of their term. We did not report the coefficient estimates of these control variables. ^c“Yes” means that we include this variable in the model.

**Table 4. Determinants of Provincial Per-Capita Industrial COD or Wastewater Discharge (2004–2015)**
	Model 3		Model 4
	Estimate (SE)	$p$ -Value	Estimate (SE)	$p$ -Value
Explanatory Variables
Decentralization
Local expenditure decentralization (H₁)	0.163	0.399	−0.417	0.050^a
	(0.193)		(0.213)
Environmental impact assessment decentralization (H $_{2 a}$ )	−0.001	0.003^a	0	0.061
	(0)		(0)
Environmental monitoring decentralization (H $_{2 b}$ )	0.094	0.003	0.126	<0.0005
	(0.031)		(0.026)
Interest groups
Industrial strength (H $_{3 a}$ )	0.218	0.209	0.073	0.587
	(0.173)		(0.134)
Foreign trade (H $_{3 b}$ )	−0.865	0.178	0.995	0.031
	(0.641)		(0.459)
One-year-lag environmental petition (H $_{3 c}$ )	0.038	0.008^a	0.027	0.007^a
	(0.014)		(0.010)
Control Variables
Provincial per-capita industrial wastewater treatment expenditure	0.078	<0.0005	0.024	0.109
	(0.021)		(0.015)
Water-use intensity (log)	−0.005	0.908	0.068	0.037
	(0.044)		(0.032)
Per-capita GDP (log)	−1.185	<0.0005	−0.779	<0.0005
	(0.059)		(0.050)
Governor characteristics^b	Yes^c		Yes
Party Secretary characteristics^b	Yes		Yes
Province fixed effects	Yes		Yes
Year fixed effects	Yes		Yes
Numbers of observations	360		360

Notes: ^aStatistical significance is being reported here as if the hypotheses are two-sided. Each noted $p -$ value is statistically significant but in the opposite direction of the originally stated hypothesis. ^bThe variables to measure Governor and Party Secretary characteristics are age, education level, and the length of their term. We did not report the coefficient estimates of these control variables. ^c“Yes” means that we include this variable in the model.

3.2. Do measures of environmental decentralization matter?

The two related hypotheses, H $_{2 a}$ and H $_{2 b}$ , aim to test whether increasing environmental decentralization, assessed in two ways, is negatively associated with water policy implementation output or outcomes. Table 3 shows that, with respect to the output dependent variable, coefficients of provincial EIA or EM decentralization do not support the original hypotheses (H $_{2 a}$ or H $_{2 b}$ ). These two measures of provincial environmental decentralization are statistically significantly positively associated with the dependent variable in both Models 1 and 2 ( $p \leq$ 0.039).

As shown in Table 4 (H $_{2 a}$ ), provincial EIA decentralization’s overall effect on COD discharge is negative and statistically significant ( $p = 0.003$ ), suggesting that increasing EIA decentralization reduces COD pollution contrary to initial expectations. However, increases in EIA and EM decentralization are both positively and significantly associated with wastewater discharge as anticipated (Model 4).

3.3. Do local interest groups matter?

Table 3 (H $_{3 a}$ ) shows that provinces where the share of industrial value-added in provincial GDP is higher tend to spend significantly more money on wastewater treatment ( $p \leq$ 0.001). These results support the initial hypothesis H $_{3 a}$ . Of the two other predictors of effect of interest groups on the output dependent variable, neither foreign trade (H $_{3 b})$ nor one-year-lag environmental petitions (H $_{3 c}$ ) are associated with per-capita provincial wastewater expenditures at an alpha level of 0.05. In assessments of water pollutants as policy implementation outcomes in Table 4, per-capita environmental petitions in the previous year are associated with more pollution in the following year ( $p \leq$ 0.008), not less as initially anticipated.

4. Discussion

This study examines the effects of fiscal decentralization, environmental decentralization, and interest groups on the outcomes and outputs of clean water policy implementation using panel data on China’s provincial administrative regions from 2004 to 2015. Table 5 includes short statements of expectations for each specific hypothesis as related to the four models, and whether the results accord with these initial expectations. These findings contribute to the discourse and debate on water governance by identifying the possible channels through which factors influence the water policy implementation in Chinese provinces.

**Table 5. Hypotheses’ Expectations and Whether Test Results Accord with Them (Yes or No) in Each of the Four Models**
			Do Test Results Accord?
Hypothesis		Expectation	Model 1	Model 2	Model 3	Model 4
H₁
Increasing fiscal decentralization	$\to$	Reduced expenditures, increasing pollution	Yes	Yes	No	No
H $_{2 a}$
Increasing decentralization of EIA	$\to$	Reduced expenditures, increasing pollution	No	No	No	Yes
H $_{2 b}$
Increasing EM decentralization	$\to$	Reduced expenditures, increasing pollution	No	No	Yes	Yes
H $_{3 a}$
Increase in industrial contributions to provincial economies	$\to$	Increased expenditures, increasing pollution	Yes	Yes	No	No
H $_{3 b}$
Greater foreign trade within provinces	$\to$	Reduced expenditures, increasing pollution	No	No	No	Yes
H $_{3 c}$
More environmental petitions per capita in the preceding year	$\to$	Increased expenditures, decreasing pollution	No	No	No	No

Note: Those that are in bold text are potentially significantly associated in the opposite direction of the original hypothesis ( $p < 0.1$ ).

This study shows that increasing fiscal decentralization tends to decrease provincial expenditures on industrial wastewater, which is largely consistent with the research conclusions of Chen et al. (2019) and Yong (2010). This may be accounted for because, as previously mentioned, local officials are under the same central government incentive structure that prioritizes economic and fiscal growth. Within this structure, cleaning up waters does not have equal priority. However, results reported here indicate that fiscal decentralization has no significant influence on water pollution as measured using per-capita COD emissions though it may be negatively associated with industrial per-capita wastewater discharge. This suggests that fiscal decentralization is not one of the main contributors to China’s industrial water pollution. This finding is inconsistent with previous studies suggesting that fiscal decentralization increases pollution based on cross-country evidence (Farzanegan and Mennel 2012; Zhang et al.2017). Although this outcome remains unexplained, we speculate that the recent shift by provincial governments towards raising a higher proportion of funds from polluting enterprises may increase the efficiency with which these funds are expended on pollution control measures (Aizawa and Yang 2010).

Second, other unanticipated findings (H $_{2 a}$ and H $_{2 b}$ ) indicate that while increasing EIA and EM decentralization at the provincial level tends to increase government’s environmental spending, increased wastewater discharge also occurs. Greater power to issue EIA plausibly increases provincial governments’ autonomy in approving new construction and increases their responsibility for environmental governance, implying more environmental spending. However, there is a negative association between COD discharge and provincial EIA decentralization (Table 4, Model 3, H $_{2 a}$ ). The unanticipated negative relationship can be partly explained by the influence of the mandatory national targets for COD discharge established in 2006. Provincial and municipal government personnel are required to meet these targets or face sanctions during cadre evaluations (Wang et al.2008). It seems plausible that provincial leaders are more eager to achieve COD discharge limits since they relate more directly to their political performance.⁵

The findings also suggest that expanding environmental monitoring power at the provincial level increases water pollutant discharge. This could occur if provincial officials misuse their power to relax pollution monitoring in high-polluting and high-profit-tax enterprises or conceal the source of environmental pollution (van Rooij 2006).

Provincial government officials may exert power over environmental monitoring department personnel through their control of funding, appointments, or promotion. This speculation may help account for the central government’s decision to implement a Central Environmental Inspection system in 2017. This system aims to monitor local authorities and their officials directly to reduce the negative influence of local protectionism on environmental policy enforcement (Li et al.2020). This and other speculation, though, require additional focused inquires.

Finally, the evidence reported here suggests that pressures from industrial groups have created positive incentives for provincial governments to increase expenditures on wastewater treatment as expected. However, neither measure of industrial water pollution was significantly influenced. In addition, foreign trade does not appear to significantly affect provincial expenditures or per-capita COD discharge but may contribute to more per-capita wastewater discharge. Scholars caution that influences of foreign trade on the environment are not consistently positive (Frankel and Rose 2005). Public pressure through petitions in the preceding year may have a weak influence, opposite to that initially anticipated, on environmental expenditures at the provincial level in the following year. This finding is inconsistent with the research conclusions of Wang and Di (2002). They suggest that public pressure has generated incentives for township governments to strengthen their efforts to enforce environmental policies. Compared to township government officials, provincial cadres may be less concerned about water pollution complaints from local residents.

These findings contribute to the research and debate on water governance and decentralization. A limitation of this study is that it focused on the implementation of clean water policies at the provincial level, but water governance mechanisms and implementation effects may differ among municipal and county-level governments within a given province. However, this information was not available. Additionally, owing to the frequent changes in statistical methods found within Chinese Statistical Yearbooks, we only had access to consistent data for measurements of industrial pollutants and provincial industrial wastewater treatment expenditures from 2004 to 2015. It would be worthwhile to explore how the variables mentioned in this research affect other types of water pollution, and how these in turn affect water policy. In China, as the central government theoretically and practically has significant power concerning environmental issues, local governments are authorized by the central government, unlike in many Western countries. In this sense, the key issue is to motivate local officials with vested interests in the current water governance system to solve practical water pollution problems rather than superficially respond to the central government’s increasingly stringent environmental regulations. Significant improvement in the water environment requires strong combined efforts from both the central and local governments.

Acknowledgments

The author would like to thank Dr. Bruce Floyd for his valuable comments on the different versions of this paper. She also acknowledges the financial support by the Chinese University of Hong Kong and the University of Auckland.

Notes

¹ A fixed-effect model is adopted over a random-effect model as the latter does not pass the Hausman test in all the regression models. For example, in Model 1, the corresponding value is 61.27, passing the significance test at the 1% significance level. The $p$ -value is less than 0.05; therefore, the fixed-effect model should be chosen.

² The estimates of the variance components and the residual scale are based on the Design Adaptive Scale estimate. This scale estimate equalizes the natural heteroscedasticity of the residuals and adjusts for the robust estimating equation for the scale itself (Koller 2016).

³ Based on the existing literature, considering the influence of GDP, the formula for measuring the environmental impact assessment is: $EIA = \frac{PEI A_{i t}}{CEI A_{t}} \times (1 - \frac{GD P_{i t}}{GD P_{t}})$ , where $PEI A_{i t}$ is the indicator of the number of cases approved in province $i$ at year $t$ . $CEI A_{t}$ is the indicator of the number of cases approved by the central government at year $t$ . ${GDP}_{i t}$ is the indicator of the GDP of province $i$ at year $t$ . ${GDP}_{t}$ is the indicator of the national GDP at year $t$ . We adjusted the measurement of environmental monitoring decentralization to follow the same method. To simplify the presentation of the variables, we do not include the formulas in Table 1.

⁴ The first one is the age of politicians, which may influence the likelihood of promotion. The Party establishes a retirement age for provincial leaders of 65 years. Thus, the center creates a glass-ceiling effect for some local politicians who are not eligible for promotion even with a strong performance (Kostka and Yu 2015). The second one is their length in the current position. When local politicians are approaching the formal end of their term, they are more likely to change their positions or retire (Landry et al.2018). Older politicians and those who have longer tenures are less likely to be promoted and, therefore, have lower incentives to be concerned about environmental issues. The third is their education level. Education is an ordered categorical variable; 0 (high school), 1 (college-level education), 2 (post-graduate degree), and 3 (doctoral degree).

⁵ It is important to acknowledge, however, that our speculation here does not account for the positive relationship between the other measure of environmental decentralization and increased COD discharge also reported here (Table 4, Models 3 and 4, H $_{2 b}$ ).

References

Aizawa, M and C Yang (2010) Green credit, green stimulus, green revolution? China’s mobilization of banks for environmental cleanup. The Journal of Environment & Development, 19(2), 119–144. Crossref, Google Scholar
Alcamo, J, P Döll, T Henrichs, F Kaspar, B Lehner, T Rösch and S Siebert (2003) Development and testing of the WaterGAP 2 global model of water use and availability. Hydrological Sciences Journal, 48(3), 317–337. Crossref, Google Scholar
Azfar, O, S Kahkonen, A Lanyi, P Meagher and D Rutherford (2018) Decentralization, governance and public services: The impact of institutional arrangements. In Devolution and Development: Governance Prospects in Decentralizing States, MS Kimenyi and P Meagher (eds.), pp. 45–88. London: Routledge. Google Scholar
Bernauer, T, R Gampfer, T Meng and Y-S Su (2016) Could more civil society involvement increase public support for climate policy-making? Evidence from a survey experiment in China. Global Environmental Change, 40, 1–12. Crossref, Google Scholar
Beyer, S (2006) Environmental law and policy in the People’s Republic of China. Chinese Journal of International Law, 5(1), 185–211. Crossref, Google Scholar
Biddle, JC and TM Koontz (2014) Goal specificity: A proxy measure for improvements in environmental outcomes in collaborative governance. Journal of Environmental Management, 145, 268–276. Crossref, Google Scholar
Cai, H, Y Chen and Q Gong (2016) Polluting thy neighbor: Unintended consequences of China’s pollution reduction mandates. Journal of Environmental Economics and Management, 76, 86–104. Crossref, Google Scholar
Chen, GC and C Lees (2018) The new, green, urbanization in China: Between authoritarian environmentalism and decentralization. Chinese Political Science Review, 3(2), 212–231. Crossref, Google Scholar
Chen, S, Y Song, Y Ding, X Qian and M Zhang (2019) Research on the strategic interaction and convergence of China’s environmental public expenditure from the perspective of inequality. Resources, Conservation & Recycling, 145, 19–30. Crossref, Google Scholar
Dombrowsky, I, N Hagemann and A Houdret (2014) The river basin as a new scale for water governance in transition countries? A comparative study of Mongolia and Ukraine. Environmental Earth Sciences, 72(12), 4705–4726. Crossref, Google Scholar
Dong, X, C Li, J Li, J Wang and W Huang (2010) A game-theoretic analysis of implementation of cleaner production policies in the Chinese electroplating industry. Resources, Conservation & Recycling, 54(12), 1442–1448. Crossref, Google Scholar
Duan, H and JV Zhan (2011) Fiscal transfer and local public expenditure in China: A case study of Shanxi Province. China Review, 11(1), 57–88. Google Scholar
Farzanegan, MR and T Mennel (2012). Fiscal decentralization and pollution: Institutions matter. Discussion Paper No. 22-2012, MAGKS Joint Discussion Paper Series in Economics. Google Scholar
Fiorino, N and R Ricciuti (2009). Interest groups and government spending in Italy. CESifo Working Paper No. 2009/2722. Available at http://hdl.handle.net/1814/42324. Google Scholar
Flörke, M, E Kynast, I Bärlund, S Eisner, F Wimmer and J Alcamo (2013) Domestic and industrial water uses of the past 60 years as a mirror of socio-economic development: A global simulation study. Global Environmental Change, 23(1), 144–156. Crossref, Google Scholar
Frankel, JA and AK Rose (2005) Is trade good or bad for the environment? Sorting out the causality. Review of Economics and Statistics, 87(1), 85–91. Crossref, Google Scholar
Han, D, MJ Currell and G Cao (2016) Deep challenges for China’s war on water pollution. Environmental Pollution, 218, 1222–1233. https://doi.org/10.1016/j.envpol.2016.08.078. Crossref, Google Scholar
Hao, Y, H Chen and Q Zhang (2016) Will income inequality affect environmental quality? Analysis based on China’s provincial panel data. Ecological Indicators, 67, 533–542. Crossref, Google Scholar
Hao, Y, Y-F Chen, H Liao and Y-M Wei (2020) China’s fiscal decentralization and environmental quality: Theory and an empirical study. Environment and Development Economics, 25(2), 159–181. Crossref, Google Scholar
He, C, F Pan and Y Yan (2012) Is economic transition harmful to China’s urban environment? Evidence from industrial air pollution in Chinese cities. Urban Studies, 49(8), 1767–1790. Crossref, Google Scholar
He, Q (2015) Fiscal decentralization and environmental pollution: Evidence from Chinese panel data. China Economic Review, 36, 86–100. Crossref, Google Scholar
Hill, M and P Hupe (2002) Implementing Public Policy: Governance in Theory and in Practice. London: Sage. Google Scholar
Hu, Y and H Cheng (2013) Water pollution during China’s industrial transition. Environmental Development, 8, 57–73. Crossref, Google Scholar
Jiang, Y and J Zheng (2017) Economic growth or environmental sustainability? Drivers of pollution in the Yangtze River Delta urban agglomeration in China. Emerging Markets Finance and Trade, 53(11), 2625–2643. Crossref, Google Scholar
Joshi, P and KA Beck (2015) Biological oxygen demand and economic growth: An empirical investigation. Water Economics and Policy, 01(02), 1550001. Link, Google Scholar
Kahn, ME, P Li and D Zhao (2015) Water pollution progress at borders: The role of changes in China’s political promotion incentives. American Economic Journal: Economic Policy, 7(4), 223–242. Crossref, Google Scholar
Koller, M (2016). Robustlmm: An R package for robust estimation of linear mixed-effects models. Journal of Statistical Software, 75(1), 1–24. Google Scholar
Koontz, TM and CW Thomas (2006) What do we know and need to know about the environmental outcomes of collaborative management? Public Administration Review, 66, 111–121. Crossref, Google Scholar
Kostka, G and AP Mol (2013) Implementation and participation in China’s local environmental politics: Challenges and innovations. Journal of Environmental Policy & Planning, 15(1), 3–16. Crossref, Google Scholar
Kostka, G and X Yu (2015) Career backgrounds of municipal party secretaries in China: Why do so few municipal party secretaries rise from the county level? Modern China, 41(5), 467–505. Crossref, Google Scholar
Kuai, P, S Yang, A Tao and ZD Khan (2019) Environmental effects of Chinese-style fiscal decentralization and the sustainability implications. Journal of Cleaner Production, 239, 118089. Crossref, Google Scholar
Lan, H, C Hart, M Yujia and ZJ Kovacic (2014). Fiscal decentralization, environmental accountability and the provision of environmental public goods in China: Evidence from 30 provinces. Working Paper. Available at http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2520956. Google Scholar
Landry, PF, X Lü and H Duan (2018) Does performance matter? Evaluating political selection along the Chinese administrative ladder. Comparative Political Studies, 51(8), 1074–1105. Crossref, Google Scholar
Li, L (2008) Political trust and petitioning in the Chinese countryside. Comparative Politics, 40(2), 209–226. Crossref, Google Scholar
Li, R, Y Zhou, J Bi, M Liu and S Li (2020) Does the Central Environmental Inspection actually work? Journal of Environmental Management, 253, 109602. Crossref, Google Scholar
Li, T, Y Han, Y Li, Z Lu and P Zhao (2016) Urgency, development stage and coordination degree analysis to support differentiation management of water pollution emission control and economic development in the eastern coastal area of China. Ecological Indicators, 71, 406–415. Crossref, Google Scholar
Li, W, J Liu and D Li (2012) Getting their voices heard: Three cases of public participation in environmental protection in China. Journal of Environmental Management, 98, 65–72. https://doi.org/10.1016/j.jenvman.2011.12.019. Crossref, Google Scholar
Lin, JY and Z Liu (2000) Fiscal decentralization and economic growth in China. Economic Development and Cultural Change, 49(1), 1–21. Crossref, Google Scholar
Ljungwall, C and M Linde-Rahr (2005). Environmental policy and the location of foreign direct investment in China. Governance Working Paper No. 22020, East Asian Bureau of Economic Research. Google Scholar
Lu, J (2021) Turnover of environmental protection officials and transboundary water pollution. Environmental Science and Pollution Research, 28(8), 10207–10223. Crossref, Google Scholar
Ma, G and J Mao (2018) Fiscal decentralisation and local economic growth: Evidence from a fiscal reform in China. Fiscal Studies, 39(1), 159–187. Crossref, Google Scholar
Ma, Y (2019) Dynamics in central–local division of the authority of EIA approval in China. Journal of Environmental Law, 31(1), 29–57. Crossref, Google Scholar
Millimet, DL (2003) Assessing the empirical impact of environmental federalism. Journal of Regional Science, 43(4), 711–733. Crossref, Google Scholar
Mol, AP and NT Carter (2006) China’s environmental governance in transition. Environmental Politics, 15(02), 149–170. Crossref, Google Scholar
Perkins, R and E Neumayer (2009) Transnational linkages and the spillover of environment-efficiency into developing countries. Global Environmental Change, 19(3), 375–383. Crossref, Google Scholar
Qi, Y, H Lu and Y Xu (2014) Research on the reform of China’s environmental decentralization system: Institutional change, quantity measurement and effect evaluation. China Industrial Economics, 1, 31–43. Google Scholar
Qi, Y and L Zhang (2014) Local environmental enforcement constrained by central–local relations in China. Environmental Policy Governance, 24(3), 216–232. Crossref, Google Scholar
Sheng, Y (2007) Global market integration and central political control: Foreign trade and intergovernmental relations in China. Comparative Political Studies, 40(4), 405–434. Crossref, Google Scholar
Sigman, H (2007). Decentralization and environmental quality: An international analysis of water pollution. NBER Working Paper No. 13098, National Bureau of Economic Research, Cambridge, MA. Google Scholar
Song, X, W Ravesteijn, B Frostell and R Wennersten (2010) Managing water resources for sustainable development: The case of integrated river basin management in China. Water Science and Technology, 61(2), 499–506. Crossref, Google Scholar
Stegarescu, D (2005) Public sector decentralisation: Measurement concepts and recent international trends. Fiscal Studies, 26(3), 301–333. Crossref, Google Scholar
Tschofen, P, IL Azevedo and NZ Muller (2019) Fine particulate matter damages and value added in the US economy. Proceedings of the National Academy of Sciences of the United States of America, 116(40), 19857–19862. Crossref, Google Scholar
van Rooij, B (2006) Implementation of Chinese environmental law: Regular enforcement and political campaigns. Development and Change, 37(1), 57–74. Crossref, Google Scholar
van Rooij, B, Q Zhu, L Na and Q Wang (2017) Centralizing trends and pollution law enforcement in China. The China Quarterly, 231, 583–606. Crossref, Google Scholar
Wang, H and W Di (2002). The determinants of government environmental performance: An empirical analysis of Chinese townships. Policy Research Working Paper, The World Bank, https://doi.org/10.1596/1813-9450-2937. Google Scholar
Wang, M, M Webber, B Finlayson and J Barnett (2008) Rural industries and water pollution in China. Journal of Environmental Management, 86(4), 648–659. https://doi.org/10.1016/j.jenvman.2006.12.019. Crossref, Google Scholar
Wang, QG, G Gu and Y Higano (2006) Toward integrated environmental management for challenges in water environmental protection of Lake Taihu Basin in China. Environmental Management, 37(5), 579–588. Crossref, Google Scholar
Wang, Y, Y Bian and H Xu (2015) Water use efficiency and related pollutants’ abatement costs of regional industrial systems in China: A slacks-based measure approach. Journal of Cleaner Production, 101, 301–310. Crossref, Google Scholar
Wu, H, Y Li, Y Hao, S Ren and P Zhang (2020) Environmental decentralization, local government competition, and regional green development: Evidence from China. Science of the Total Environment, 708, 135085. Crossref, Google Scholar
Wu, J, M Xu and P Zhang (2018) The impacts of governmental performance assessment policy and citizen participation on improving environmental performance across Chinese provinces. Journal of Cleaner Production, 184, 227–238. Crossref, Google Scholar
Xie, L (2016) Environmental governance and public participation in rural China. China Information, 30(2), 188–208. Crossref, Google Scholar
Yew, WL and Z Zhu (2019) Innovative autocrats? Environmental innovation in public participation in China and Malaysia. Journal of Environmental Management, 234, 28–35. Crossref, Google Scholar
Yong, F (2010) Fiscal decentralization, governance and non-economic public goods provision. Economic Research Journal, 8, 4–14 (in Chinese). Google Scholar
You, D, Y Zhang and B Yuan (2019) Environmental regulation and firm eco-innovation: Evidence of moderating effects of fiscal decentralization and political competition from listed Chinese industrial companies. Journal of Cleaner Production, 207, 1072–1083. Crossref, Google Scholar
Zhang, C (2017) Political connections and corporate environmental responsibility: Adopting or escaping? Energy Economics, 68, 539–547. Crossref, Google Scholar
Zhang, C, J Chen and Z Wen (2012) Assessment of policy alternatives and key technologies for energy conservation and water pollution reduction in China’s synthetic ammonia industry. Journal of Cleaner Production, 25, 96–105. Crossref, Google Scholar
Zhang, J and X Fu (2008) FDI and environmental regulations in China. Journal of the Asia Pacific Economy, 13(3), 332–353. Crossref, Google Scholar
Zhang, K, Z-Y Zhang and Q-M Liang (2017) An empirical analysis of the green paradox in China: From the perspective of fiscal decentralization. Energy Policy, 103, 203–211. Crossref, Google Scholar
Zhang, L, L Xiong, B Cheng and C Yu (2018) How does foreign trade influence China’s carbon productivity? Based on panel spatial lag model analysis. Structural Change and Economic Dynamics, 47, 171–179. Crossref, Google Scholar
Zhou, K-Y and WR Sheate (2011) EIA application in China’s expressway infrastructure: Clarifying the decision-making hierarchy. Journal of Environmental Management, 92(6), 1471–1483. Crossref, Google Scholar
Zhu, H, Q Fu and Q Wei (2014) Calculation of efficiency on environmental expenditure and study on influential factors. China Population, Resources and Environment, 6, 91–96. Google Scholar