Assessing Environmental, Social, and Governance Risks in the Water, Energy, Land, and Food Nexus, Towards a Just Transition to Sustainable Energy in China

Abstract

The transition to sustainable energy in China is closely intertwined with environmental, social, and governance (ESG) risks within the water–energy–land–food (WELF) nexus. This study examines the complex interdependencies among these resources and evaluates the ESG challenges that may hinder or accelerate the energy transition. By integrating policy analysis and quantitative risk assessment, this research identifies key ESG risks, such as water scarcity, land-use conflicts, food security concerns, and social equity issues. The findings highlight the need for holistic governance frameworks and cross-sectoral strategies to mitigate ESG risks while ensuring a resilient and just energy transition. This study provides policy recommendations for aligning energy development with sustainable resource management, contributing to China’s long-term climate and economic goals.

1. Introduction

The global shift towards sustainable energy is essential for tackling climate change, strengthening energy security, and fostering economic resilience [1,2]. Nations around the world are working to lower carbon emissions and transition to renewable energy sources, in line with global initiatives like the Paris Agreement and the United Nations Sustainable Development Goals (SDGs) [3,4]. However, this transition presents significant challenges, especially in managing the interconnected relationships among vital resources, including water, energy, land, and food [5,6]. The water–energy–land–food (WELF) nexus is a critical system where the management and consumption of one resource directly affect the others, resulting in complex sustainability trade-offs [7,8]. As the demand for clean energy continues to rise, ensuring an equitable and balanced transition requires addressing the environmental, social, and governance (ESG) risks associated with this nexus [9].

China, as the world’s largest energy consumer and carbon emitter, plays a pivotal role in shaping the global energy landscape [10]. The country has made ambitious commitments to peak carbon emissions by 2030 and achieve carbon neutrality by 2060, signaling a major transformation of its energy system [11,12,13]. However, China’s transition to low-carbon energy is deeply entangled with ESG risks related to resource constraints, biodiversity loss, and social equity concerns [14,15]. Large-scale renewable energy projects, such as hydropower, solar farms, and bioenergy, require significant land and water resources, which can disrupt ecosystems and compete with food production [16,17]. Additionally, the extraction and processing of critical minerals essential for clean energy technologies, such as lithium and rare earth elements, raise concerns about environmental degradation and human rights violations [18].

Given these challenges, a comprehensive assessment of ESG risks within the WELF nexus is crucial to ensuring a resilient and equitable energy transition [19,20]. This study examines the ESG implications of China’s energy transition through the lens of the WELF nexus, highlighting key environmental, economic, and governance risks. By integrating policy analysis, risk assessment, and case studies, this research identifies potential trade-offs and synergies in resource management [21,22]. The findings offer insights supporting policymakers, businesses, and stakeholders to develop holistic strategies that balance energy security, sustainability, and social well-being, ultimately contributing to both China’s national goals and global climate commitments [23]. China’s rapid industrialization and urbanization have intensified competition for water, land, and food resources, exacerbating ESG risks that could hinder sustainable energy progress [24,25]. For example, large-scale renewable energy projects such as hydropower and bioenergy require extensive water and land resources, raising concerns about ecosystem disruption and agricultural displacement [26,27]. Similarly, the expansion of lithium and rare earth mining, essential for clean energy technologies, poses governance challenges related to environmental degradation and social justice [28,29]. As China accelerates its shift toward renewable energy, a comprehensive understanding of ESG risks within the WELF nexus is necessary to ensure a just and resilient energy transition [30,31,32].

This paper is structured into four sections following this introduction. The methodology and data sources are outlined in Section 2. Section 3 presents the findings, covering the distribution of ESG risks and the effects of energy transition policy on ESG risks within the water–energy–land–food nexus. Section 4 explores discussions and implications, while Section 5 concludes the paper.

2. Materials and Methods

2.1. Conceptualization of ESG Risks in the WELF Nexus

The WELF nexus represents the interconnectedness of critical resources—water, energy, land, and food—which are deeply integrated in shaping both environmental sustainability and human well-being. Within this nexus, ESG risks emerge as interconnected factors that affect sustainable development across these domains. A clear understanding of ESG risks in the WELF nexus requires a conceptual breakdown of each type of risk and their interrelationships.

Environmental risk (E): In the WELF nexus, environmental risks arise from the depletion, contamination, or mismanagement of natural resources, impacting water, energy, land, and food systems [3]. For example, over-extraction of water for energy production or agricultural use can lead to water scarcity, while land degradation can disrupt food production. Social risk (S): Social risks within the WELF nexus relate to the human and community impacts of how water, energy, land, and food resources are managed and distributed [10,11]. These risks include issues of equity, access, and social justice—such as the unequal distribution of energy or water resources, the displacement of communities due to land-use changes, or the health and livelihood consequences of poor environmental practices [4,5,6]. Governance risk (G): Governance risks are tied to the management and regulation of the water, energy, land, and food systems within a region. In the WELF nexus, governance risks manifest when policies, regulations, or institutional frameworks are weak, inefficient, or poorly implemented [26,27]. Poor governance can exacerbate environmental degradation, worsen social inequalities, and lead to ineffective responses to crises like resource shortages or climate change impacts [30].

2.2. Principal Component Analysis and Comprehensive Index Evaluation Method

When evaluating regional ESG risks in the WELF nexus, the first step is to use the range standardization method to standardize the original data, and then use the principal component analysis method to redesign the dimensionality and obtain the weights of the main variables (Figure 1). For details, please see Appendix A.

Variable Standardization Processing

Considering that the dimensions of different indicators are different and cannot be directly compared, the indicators need to be standardized first.

$$\mathrm{Positive} \mathrm{indicators}: X_{ij}={\displaystyle \frac{x_{ij}-x_{min}}{x_{max}-x_{min}}}$$

(1)

$$\mathrm{Negative} \mathrm{indicators}: X_{ij}={\displaystyle \frac{x_{max}-x_{ij}}{x_{max}-x_{min}}}$$

(2)

2.3. Principal Component Analysis

Assuming there are n regions and p evaluation indicators; if the original indicators are denoted as $x_p$, and their principal components after dimensionality reduction are denoted as $y_m(\mathrm{m}\le \mathrm{p}$), we can define the following:

$$\left\{\begin{array}{c}{y}_1={\beta }_{11}{x}_1+{\beta }_{12}{x}_2+...+{\beta }_{1p}{x}_p\\ y_2={\beta }_{21}{x}_1+{\beta }_{22}{x}_2+...+{\beta }_{2p}{x}_p\\ .\\ .\\ .\\ y_m={\beta }_{m1}{x}_1+{\beta }_{m2}{x}_2+...+{\beta }_{mp}{x}_p\end{array}\right.$$

(3)

Then, the basic calculation process of principal component analysis is as follows.

First, set the original estimation matrix $\mathrm{X}={\left({\mathrm{x}}_{\mathrm{i}\mathrm{j}}\right)}_{\mathrm{n}\times \mathrm{p}}$, where ${\mathrm{x}}_{\mathrm{i}\mathrm{j}}$ represents the standardized data of indicator j for region i. At the same time, calculate the covariance matrix for each variable.

Next, based on the covariance matrix $R_{ij}$, calculate the eigenvalues ${\lambda }_j$, variance contribution rate ${\sigma }_j$, and cumulative variance contribution rate ${\alpha }_j$. According to the principle that eigenvalues should be greater than 1 and the cumulative variance contribution rate should be greater than 70%, select the first m variables ${\mathrm{y}}_1,{\mathrm{y}}_2\dots {\mathrm{y}}_{\mathrm{m}}$ that meet the criteria as principal components to replace the original p series variables:

$${\sigma }_j={\lambda }_j/{\sum }_{p=1}^p{\lambda }_j$$

(4)

$${\alpha }_j={\sum }_{j=1}^m{\lambda }_j/{\sum }_{j=1}^p{\lambda }_j$$

(5)

Subsequently, the initial factor loading matrix $a_{mj}$ is determined based on the correlation coefficient between the principal component ${\mathrm{y}}_{\mathrm{m}}$ and the initial index $x_j$ to explain the degree of correlation between the principal component and each evaluation index, and further obtain the principal component coefficient matrix ${\gamma }_{mj}=a_{mj}/\sqrt{{\lambda }_j}$ with each principal component score $y_m={\gamma }_{jm}^T{x}_j$.

Finally, based on the ratio of the variance contribution rate of each principal component to the cumulative variance contribution rate of each principal component, the weight ${\theta }_j$ of each principal component is calculated.

Calculate ESG index risks:

$${ESG}_{it}=y_{tij}\times {\theta }_j$$

(6)

2.4. Three-Dimensional Kernel Density Estimation

This paper employs the three-dimensional kernel density estimation method to model ESG risk data across various provinces in China, analyzing distribution patterns and evolutionary trends from 2007 to 2022. Assuming $\mathrm{f}\left(\mathrm{x},\mathrm{y}\right)$ represents the joint kernel density estimation function of the two-dimensional random variables $(\mathrm{X},\mathrm{Y})$, where the random variables $\mathrm{X}=({\mathrm{x}}_1,{\mathrm{x}}_2,...{\mathrm{x}}_{\mathrm{m}})$ and $\mathrm{Y}=({\mathrm{y}}_1,{\mathrm{y}}_2,...{\mathrm{y}}_{\mathrm{n}})$ follow independent distributions, the specific formula is as follows:

$$f\left(x,y\right)={\displaystyle \frac{1}{n{h}_x{h}_y}}{\sum }_{i=1}^n{K}_x{\displaystyle \frac{x_i-\overline{x}}{h_x}}{K}_y{\displaystyle \frac{y_i-\overline{y}}{h_y}}$$

(7)

where ${\displaystyle \frac{K_x(x_i-\overline{x})}{h_x}}$、${\displaystyle \frac{K_y(y_i-\overline{y})}{h_y}}$ are kernel functions, and h represents the bandwidth of the kernel density function, determining the smoothness of the kernel density curve and the estimation accuracy.

2.5. Baseline Regression Model

In order to explore the impact of energy transition policies on regional ESG risks in the WELF nexus, a double-difference method is employed, with provincial ESG risk indices as the dependent variables. The construction involves creating interaction terms between pilot reforms in energy use rights and virtual variables before and after policy implementation. These are then utilized as explanatory variables of ESG risks. The pilot scheme outlined in the Pilot Scheme for Compensated Use and Trading System of Energy Use Rights issued by the National Development and Reform Commission (NDRC) is regarded as a quasi-natural experiment. Provinces implementing pilot programs for energy transition policies, such as Zhejiang, Fujian, Henan, and Sichuan, are designated as the experimental group, while other regions serve as the control group. Moreover, the policy release year of 2016 is designated as the shock time point. Empirical testing is conducted to examine whether energy transition policies are conducive to enhancing regional ESG risks in WELF nexus indices. The specific model is as follows:

$$\begin{gathered} {ESG}_{it}={\beta }_0+{\beta }_1{Time}_t\times {Treat}_i+{\beta }_{2}ln{pgdp}_{it}+{\beta }_{3}ln{indus}_{it}+{\beta }_{4}ln{urban}_{it}+{\beta }_{5}ln{market}_{it} \\ +{\beta }_{6}ln{open}_{it}+{\beta }_{6}ln{er}_{it}+{\epsilon }_{it} \end{gathered}$$

(8)

In the equation, i represents provinces and t denotes years; ESG stands for the ESG risk index of each region; ${pgdp}_{it}$ represents the economic development level of province i in year t; ${indus}_{it}$ denotes the level of industrialization in province i in year t; ${urban}_{it}$ represents the urbanization rate of province i; ${market}_{it}$ indicates the level of marketization in province i during year t; ${Open}_{it}$ represents the degree of openness of province i; ${er}_{it}$ denotes the level of environmental regulation in province i; ${Treat}_i$ is a virtual variable, where if province i is part of the pilot reform for energy use rights, ${Treat}_{it}=1$, and otherwise, ${Treat}_{it}=0$; ${Time}_t$ is a time dummy variable. Its value is 0 before the introduction of the energy transition policy, and 1 in the year of policy implementation and thereafter.

2.6. Mediating Effect Model

To further investigate how energy transition policies influence regional ESG risks within the WELF nexus, empirical tests are conducted to examine the mediating effects of technological progress, industrial structure upgrading, and resource endowment dependence. The constructed model is as follows:

$$Q_{it}={\alpha }_0+{\beta }_1{Time}_t\times {Treat}_i+{\sum }_{n=1}^6{\gamma }_n{Z}_{nit}+{\mu }_i+v_t+{\epsilon }_{it}$$

(9)

$${ESG}_{it}={\alpha }_0+{\beta }_1{Time}_t\times {Treat}_i+{\beta }_2{Q}_{it}+{\sum }_{n=1}^6{\gamma }_n{Z}_{nit}+{\mu }_i+v_t+{\epsilon }_{it}$$

(10)

In this equation, $Q_{it}$ represents the mediating variables, including technological progress, industrial structural upgrading, and resource endowment dependence. Other symbols have the same meanings as in Equation (8).

2.7. Data Sources and Variables

The dependent variable is the ESG risk index (ESG). Drawing from the conceptual framework of ESG risks and considering data availability, a regional ESG risk evaluation system for the WELF nexus is developed. This system includes 15 indicators encompassing environmental, social, and governance dimensions (see

Table 1. ESG risk evaluation index system.

Target	Dimensions	Definition	Indicators	Positive/Negative	Sources
ESG risks	Environmental system	Potential negative impacts of human activities on the natural environment within a specific region.	Sulfur dioxide emissions per unit of energy consumption	Negative	[33,34]
			Smoke (powder) dust emissions per unit of energy consumption	Negative	[35]
			Nitrogen oxide emissions per unit of energy consumption	Negative	[36,37,38]
			Energy consumption low-carbonization index	Positive	[39,40]
			Per-capita water resource stock	Positive	[10,11]
	Social system	Societal impacts that can arise from economic, political, and environmental actions at the regional level.	Energy consumption per capita	Negative	[20]
			Energy consumption growth rate	Negative	[33,34]
			Contribution of clean energy consumption to employment	Positive	[2,3]
			Contribution of clean energy consumption to the economy	Positive	[5]
			Coal energy consumption share	Negative	[16,17]
	Governance system	Potential for weak or ineffective governance structures to negatively affect the region’s economic and social outcomes in the WELF nexus.	The degree of agglomeration of clean energy consumption	Positive	[18,19]
			Government investment in environmental protection	Positive	[41,42]
			Coal reserves	Positive	[33,36]
			Growth rate of total primary energy consumption	Positive	[22,35]
			Energy efficiency	Positive	[43]

). The ESG risk index for each region is calculated using principal component analysis and comprehensive index evaluation methods. The data for this study cover 30 provinces and municipalities in mainland China (excluding Hong Kong, Macau, Taiwan, and Tibet). The main sources of data are the China Statistical Yearbook from 2008 to 2023, as well as various provincial statistical yearbooks and development statistical bulletins from different years (

Table 2. Descriptive statistics of core variables.

Variable Symbol	Variable Meaning	Sample Size	Mean	Standard Deviation	Minimum	Maximum
ESG	ESG risks	480	0.000	0.994	−3.252	2.537
lntec	Technology improvement	480	9.903	1.627	5.407	13.895
lnupi	Industrial structural upgrade	480	1.216	0.037	1.142	1.344
lnred	Resource endowment dependence	480	2.347	1.072	0.095	4.644
lnpgdp	Economic development level	480	10.661	0.588	8.967	12.158
lnindus	Industrialization level	480	0.297	0.064	0.095	0.453
lnurban	Urbanization rate	480	4.045	0.227	3.376	4.506
lnmarket	Marketization level	480	2.149	0.224	1.472	2.595
lnopen	Level of openness to the outside world	480	1.814	0.812	0.570	6.661
lner	Intensity of environmental regulation	480	0.362	0.304	0.000	1.277

).

The primary explanatory variables include an interaction term between the policy implementation timeline (Time) and the policy participation indicator (Treat), denoted as (Time × Treat). The Treat variable is a binary indicator, taking a value of 1 for pilot regions (Zhejiang, Fujian, Henan, and Sichuan) and 0 for non-pilot regions.

The energy transition policy, launched in 2016 by the National Development and Reform Commission (NDRC) under the Pilot Scheme for Compensated Use and Trading System of Energy Use Rights, requires pilot regions to cap energy consumption and allocate initial energy use rights per unit. This initiative is designed to regulate demand and reduce environmental impact. The Time variable, also binary, is assigned 0 for years before 2016 and 1 for 2016 and later, signifying the policy’s rollout and expansion.

Several control variables are incorporated to account for regional differences. These include the economic development level (pgdp), measured by per capita GDP; industrialization level (indus), captured by the share of industrial value added in GDP; urbanization rate (urban), reflected by the proportion of the urban population; marketization level (market), based on the marketization index from the Fan Gang research group; degree of openness (open), determined by the ratio of total trade to GDP; and environmental regulation intensity (er), assessed through an index incorporating industrial wastewater discharge, industrial SO₂ emissions, and industrial smoke emissions, where higher values indicate weaker regulatory enforcement.

Additionally, three mediating variables are considered: technological progress (tec), industrial structural upgrading (upi), and resource endowment dependence (red). Technological progress is quantified by the number of domestic patent approvals. Industrial structural upgrading is evaluated through the ratio of tertiary to secondary industry output. Resource endowment dependence is represented by the number of urban employees per mining unit.

3. Results

3.1. Temporal Evolution Trend of ESG Risks in the WELF Nexus

Figure 2 presents the three-dimensional kernel density estimation and heatmap of regional ESG risks in the WELF nexus from 2007 to 2022, enabling the assessment of spatial distribution and temporal evolution trends of ESG risks during the study period.

From 2007 to 2022, the ESG risk index displayed a consistent upward trajectory, indicating an overall improvement in environmental, social, and governance aspects of energy management across China. However, it is noteworthy that in 2022, there was a slight dip in the ESG risk index, suggesting a potential deviation from the upward trend, which warrants further investigation into the factors contributing to this decrease.

Significant regional disparities exist in ESG risks across China. Certain regions such as Beijing, Hainan, and Qinghai demonstrate notably higher ESG risk indices compared to the national average. These regions likely benefit from favorable environmental policies, advanced infrastructure, and proactive governance strategies. On the other hand, regions like Shanxi, Inner Mongolia, Guizhou, and Shaanxi exhibit considerably lower ESG risk indices, indicating potential challenges related to environmental degradation, social inequalities, and governance deficiencies. Understanding these regional variations is crucial for targeted policy interventions and resource allocation to address disparities and promote balanced development across the country.

China’s ESG risk landscape reflects substantial internal differences and strong agglomeration effects. While the overall trend suggests a gradual convergence of ESG risk indices among different regions over the years, significant disparities persist. The observed fluctuations in regional differences underscore the complex interplay of factors influencing energy management practices, including regional economic structures, resource endowments, policy frameworks, and socio-environmental conditions. Efforts to mitigate these disparities and enhance overall ESG risk performance require comprehensive policy approaches tailored to the unique challenges and opportunities within each region, along with concerted efforts to promote collaboration, knowledge sharing, and capacity-building across the country.

3.2. Spatial–Temporal Distribution of ESG Risks in the WELF Nexus

From 2007 to 2022, the average ESG risk outcomes exhibited a notable upward trajectory, indicating a positive trend in environmental, social, and governance aspects of energy management over the years. The average ESG risk outcomes increased from −0.8032 in 2007 to 0.5750 in 2022, representing a significant improvement. By 2021, the ESG risk outcomes had nearly doubled relative to 2007, underscoring the substantial progress in energy sustainability practices (Figure 3).

Across the period spanning from 2007 to 2022, the average ESG risks remained relatively stable at 0.0000. However, there were noticeable fluctuations in annual averages, with the highest recorded in 2021 (0.6170) and the lowest in 2007 (−0.8032), indicating varying levels of energy sustainability performance over time.

An examination of yearly ESG indices from 2007 to 2022 reveals distinct patterns. The years with the highest ESG indices were 2021 (0.6170), 2022 (0.5750), 2020 (0.5567), 2019 (0.4800), and 2018 (0.4365). Notably, 43.75% of the years had ESG indices above the mean, while 56.25% fell below the mean, highlighting the variability in energy sustainability performance across different years. Conversely, the years with the lowest ESG indices were 2007 (−0.8032), 2008 (−0.6869), 2009 (−0.6125), 2010 (−0.3848), and 2011 (−0.2416), indicating periods of relatively lower energy sustainability performance.

Analyzing ESG risks across the 30 provinces and municipalities reveals significant disparities. For instance, Beijing exhibited the highest ESG risk index (1.7058), followed by Hainan (1.5094), Zhejiang (1.2309), Guangdong (1.2123), and Shanghai (1.1275). Conversely, Shanxi had the lowest ESG risk index (−2.2785), followed by Inner Mongolia (−1.6724), Ningxia (−1.1453), Hebei (−0.9155), and Guizhou (−0.7414). This disparity underscores the diverse energy sustainability performance among different regions, necessitating targeted interventions to address the underlying challenges and promote balanced development across the country.

3.3. Impact of Energy Transition Policy on ESG Risks in the WELF Nexus

The findings suggest that prior to 2016, there was no discernible gap in the growth trajectories of ESG risks between regions designated as pilot areas and those not included in the pilot program. This indicates that, in the pre-implementation phase, both pilot and non-pilot regions experienced similar trends in ESG risk development.

However, following the implementation of the policy during the period from 2016 to 2022, all key explanatory variables exhibited statistically significant positive coefficients. This implies that subsequent to policy enactment, differences emerged in the ESG risk indices between pilot provinces and non-pilot provinces. The positive coefficients suggest that the policy intervention had a notable impact on energy sustainability outcomes, indicating that the energy transition policy effectively stimulated changes in the energy structure of pilot provinces.

Furthermore, verifying the parallel trend assumption is crucial for assessing the model’s reliability. The results indicate that before policy implementation, ESG risk trends in both pilot and non-pilot regions were comparable. This supports the appropriateness of the difference-in-differences model, reinforcing the credibility of the analytical approach and strengthening the robustness of findings on the energy transition policy’s impact on energy sustainability (Figure 4).

The implementation of energy transition policies represents a pivotal step in bolstering the regional energy environmental, social, and governance (ESG) standards. These policies are designed to facilitate the transition towards cleaner and more sustainable energy practices, thereby fostering improvements in various aspects of energy management, including environmental protection, social equity, and governance efficiency. By implementing these policies, governments can effectively elevate the overall ESG risk performance within their jurisdictions, ensuring a more sustainable and resilient energy landscape for future generations.

It is observed that an increase in development level, marketization level, openness to foreign markets, and environmental regulation intensity correlates positively with an augmentation in ESG risk outcomes. This suggests that regions experiencing higher levels of economic development, market liberalization, international engagement, and stringent environmental regulations tend to exhibit better ESG risk outcomes. These factors contribute to fostering a conducive environment for sustainable energy practices, promoting investments in renewable energy, enhancing energy efficiency measures, and mitigating environmental risks associated with energy production and consumption.

Conversely, increases in industrialization level and urbanization rate may exert a downward pressure on ESG risks. As regions undergo rapid industrialization and urban expansion, they often face challenges such as increased energy consumption, higher emissions levels, environmental degradation, and social disparities. These factors can undermine the overall sustainability of energy systems, leading to lower ESG risk scores. Hence, policymakers need to implement targeted measures to address these challenges, such as promoting cleaner production technologies, implementing urban planning strategies that prioritize energy efficiency and environmental sustainability, and enhancing social inclusion initiatives to ensure equitable access to clean energy resources and benefits (

Table 3. Baseline regression results.

Variables	Dependent Variable: ESG Risks
	(1)	(2)	(3)	(4)
Treat × Time	0.547 ***	0.707 ***	0.422 ***	0.372 ***
	(0.146)	(0.123)	(0.112)	(0.105)
lnpgdp		0.748 ***	0.878 ***	1.117 ***
		(0.096)	(0.153)	(0.143)
lnindus		−6.199 ***	−7.667 ***	−4.884 ***
		(0.493)	(0.452)	(0.508)
lnurban			−1.484 ***	−2.595 ***
			(0.282)	(0.285)
lnmarket			1.935 ***	1.877 ***
			(0.189)	(0.177)
lnopen				0.135 ***
				(0.050)
lner				−0.881 ***
				(0.104)
Cons	−0.136	−5.722 ***	−4.945 ***	−3.601 ***
	(0.138)	(0.989)	(0.878)	(0.824)
N	480	480	480	480
R2	0.498	0.655	0.733	0.777
Year fixed effects	Control	Control	Control	Control
Region fixed effects	Control	Control	Control	Control

Note: *** indicates that they have passed the significance test at the 1% levels; the values in parentheses are robust standard errors.

).

3.4. Robustness Test

The energy transition policy, launched in 2016, designated four pilot regions as the experimental group, with the remaining provinces serving as the control group. To ensure comparability, six observable variables—economic development level, industrialization level, urbanization rate, marketization level, openness to foreign markets, and environmental regulation intensity—were used for matching. The effectiveness of this matching process is demonstrated in

Table 4. Balance test of each variable before and after propensity value matching.

Variable	Before and After Matching (BM/AM)	Experimental Group	Control Group	Deviation	Deviation Change Range (%)
lnpgdp	BM	10.776	10.643	23.200
	AM	10.843	10.821	3.800	83.600
lnindus	BM	0.329	0.292	68.900
	AM	0.325	0.323	2.600	96.200
lnurban	BM	4.013	4.050	−16.800
	AM	4.046	4.036	4.400	73.600
lnmarket	BM	2.297	2.126	94.500
	AM	2.311	2.307	2.600	97.200
lnopen	BM	1.703	1.831	−18.300
	AM	1.777	1.751	3.800	79.300
lner	BM	0.477	0.344	50.700
	AM	0.467	0.468	−0.600	98.800

and Figure 5.

The results indicate that, considering self-selection bias, the implementation of the energy transition policy significantly enhances the regional ESG risk outcomes in WELF nexus levels, consistent with the benchmark regression results. This underscores the robustness and reliability of the findings, providing evidence of the policy’s effectiveness in improving energy sustainability outcomes (Table 4 and

Table 5. PSM-DID estimation results.

Variable	(1)	(2)
Treat × Time	0.545 ***	0.396 ***
	(0.156)	(0.111)
Cons	−0.130	−3.557 ***
	(0.140)	(0.833)
N	480	480
R2	0.496	0.776
Control variables	No	Yes
Year fixed effects	Control	Control
Region fixed effects	Control	Control

Note: *** indicates that they have passed the significance test at the 1% levels; the values in parentheses are robust standard errors.

).

3.5. Regional Inequity: Impact of Energy Transition Policy in the WELF Nexus

Recognizing that regional disparities in resource endowments may affect the impact of energy rights on ESG risks, this study categorizes the sample into eastern, central, and western regions based on varying economic development levels. The detailed results are provided in

Table 6. Heterogeneity analysis.

Variable	Dependent Variable: ESG Risks
	East (1)	Middle and West (2)
Treat × Time	0.657 ***	0.410 ***
	(0.144)	(0.142)
Control variable	Yes	Yes
con	−3.909 ***	−2.482 *
	(1.180)	(1.315)
N	176	304
R2	0.800	0.710

Note: *, and *** indicates that they have passed the significance test at the 10%, and 1% levels, respectively.

.

The observed increase of 0.657 units in ESG risk outcomes for the eastern regions and 0.410 units for the central and western regions following the implementation of the energy transition policy underscores the policy’s positive impact on energy sustainability outcomes. This indicates that the policy intervention has effectively contributed to enhancing ESG risk outcome levels across all regions, albeit to varying degrees.

Furthermore, when comparing the effects of the energy transition policy on ESG risks across different regions, it becomes apparent that the policy has a more pronounced promoting effect on ESG risk outcomes in the eastern regions compared to the central and western regions. This discrepancy could be attributed to several factors, including variations in regional economic structures, industrial compositions, resource endowments, and policy implementation effectiveness.

Eastern regions, typically characterized by more developed economies, higher levels of industrialization, and greater access to advanced technologies and resources, may have been better positioned to leverage the benefits of the energy transition policy. Additionally, these regions may have had more robust governance mechanisms and institutional frameworks in place to facilitate policy implementation and enforcement, leading to a more significant improvement in ESG risk outcomes.

On the other hand, central and western regions, which often face challenges such as resource constraints, lower levels of economic development, and infrastructural limitations, may have experienced a slightly lesser impact from the policy implementation. Nonetheless, the observed positive effect on ESG risks in these regions highlights the policy’s potential to contribute to energy sustainability improvements nationwide, albeit with varying magnitudes across different regional contexts.

3.6. Driving Factors of Energy Transition Policy on ESG Risks in the WELF Nexus

To further examine how the energy transition policy influences ESG risks, an empirical analysis was conducted using the intermediate effect model, as specified in Equations (9) and (10). The results, presented in

Table 7. Estimated path results of the impact of energy structure transformation policies on ESG risks.

Variable	lntec	lnupi	lnred	ESG	ESG	ESG
	(1)	(2)	(3)	(4)	(5)	(6)
Treat × Time	0.324 *	0.014 **	−0.499 ***	0.529 ***	0.612 ***	0.347 ***
	(0.173)	(0.007)	(0.178)	(0.107)	(0.141)	(0.099)
lntec				0.247 ***
				(0.041)
lnupi					1.993 ***
					(0.317)
lnred						−0.302 ***
						(0.026)
Control variable	Yes	Yes	Yes	Yes	Yes	Yes
con	−9.373 ***	1.215 ***	8.700 ***	−3.402 ***	−4.870 ***	−2.469 ***
	(1.366)	(0.002)	(1.381)	(0.879)	(0.764)	(0.789)
N	480	480	480	480	480	480
R2	0.755	0.534	0.433	0.762	0.538	0.803

Note: *, **, and *** indicates that they have passed the significance test at the 10%, 5%, and 1% levels, respectively.

, reveal the following key findings.

The energy transition policy plays a pivotal role in enhancing regional ESG risk outcomes in the WELF nexus through various mechanisms. Firstly, by promoting technological progress, the policy encourages the adoption of cleaner and more efficient energy technologies, which in turn reduces environmental impacts and improves energy sustainability. Additionally, the policy facilitates industrial structural upgrading by incentivizing the transition towards cleaner and more sustainable industrial processes and practices. This shift towards cleaner industries leads to reduced emissions and resource consumption, contributing to improved ESG risk outcomes. Moreover, the policy helps mitigate resource endowment dependency by diversifying energy sources and promoting renewable energy adoption. This reduces reliance on finite and environmentally damaging resources, further enhancing ESG risk outcomes.

Regions that experience faster technological progress, higher levels of industrial structural upgrading, and weaker dependency on resource endowments are likely to exhibit higher ESG risk outcomes. This is because advancements in technology enable the adoption of cleaner and more efficient energy solutions, while industrial structural upgrading promotes the transition towards cleaner production processes and industries. Additionally, regions with weaker resource endowment dependency are better positioned to adopt sustainable energy practices and diversify their energy sources, thereby improving ESG risk outcomes.

To further enhance regional ESG outcomes in the WELF nexus, interventions should prioritize strategies that foster technological progress and industrial structural upgrading while simultaneously reducing the reliance on resource endowments. This could involve investing in research and development initiatives to promote innovation in clean energy technologies, providing incentives for industries to adopt cleaner production processes, and implementing policies that encourage energy diversification and sustainability practices.

Overall, the implementation of the energy transition policy serves as a catalyst for facilitating technological progress and industrial structural upgrading while reducing the reliance on resource endowments. By addressing these key aspects, the policy contributes significantly to the elevation of ESG risk outcomes in different regions, ultimately promoting sustainable energy development and environmental conservation.

3.7. Dynamic Effect of Energy Transition Policy on ESG Risks in the WELF Nexus

We explored the dynamic marginal effects of the energy transition policy on regional ESG risks within the WELF nexus, and dummy variables for each year were included in the analysis. The detailed results are presented in

Table 8. Dynamic effects of energy transition policies on ESG risks.

Variable	ESG Risks
Treat * 2016	0.740 *
	(0.414)
Treat * 2017	0.874 **
	(0.414)
Treat * 2018	0.971 **
	(0.414)
Treat * 2019	1.052 **
	(0.414)
Treat * 2020	1.095 ***
	(0.414)
Treat * 2021	1.112 ***
	(0.414)
Treat * 2022	1.077 ***
	(0.414)
CONS	0.601 ***
	(0.063)
R2	0.759
Control variable	Yes
N	480

Note: *, **, and *** indicates that they have passed the significance test at the 10%, 5%, and 1% levels, respectively.

.

The implementation of the energy transition policy continues to have a sustained promoting effect on regional ESG risk outcomes in the WELF nexus. This implies that the policy’s impact persists over time, contributing to ongoing improvements in energy sustainability across regions.

The year 2021 stands out as having the most significant promoting effect of the energy transition policy on regional ESG risk outcomes in the WELF nexus. This suggests that the policy’s effectiveness reached its peak in 2021, resulting in the most substantial improvements in energy sustainability outcomes during that year. These findings underscore the importance of continued policy support and implementation in fostering positive changes in energy sustainability, with 2021 representing a particularly noteworthy period of heightened effectiveness for the energy transition policy in enhancing regional ESG risk outcomes in the WELF nexus.

4. Discussion

The discussion of ESG risks and driving factors in the context of energy transition policies at the provincial level in China illuminates the intricate dynamics shaping the trajectory of sustainable development in one of the world’s largest economies. This section delves into key insights gleaned from the analysis, highlighting the implications for policy, governance, and stakeholder engagement [44,45,46].

The analysis reveals the interconnected nature of ESG risks, underscoring the need for a holistic approach to energy transition policymaking [47,48]. Environmental risks, ranging from air pollution to water scarcity, pose immediate threats to public health and ecological integrity, necessitating stringent regulatory measures and technological interventions [49,50]. Social risks, such as land acquisition conflicts and community opposition, underscore the importance of inclusive decision-making processes and meaningful stakeholder engagement [51,52,53]. Governance risks, including regulatory capture and policy inertia, highlight the challenges of institutional capacity-building and the imperative of transparent and accountable governance mechanisms [54,55,56].

This Discussion elucidates the significance of regional disparities and contextual specificities in shaping energy transition policies across different provinces in China. While coastal provinces with higher levels of economic development may prioritize renewable energy deployment and emissions reduction targets, inland provinces with abundant coal resources may face greater challenges in transitioning away from fossil fuels [57,58,59]. Moreover, provinces with a heavy reliance on energy-intensive industries may confront social and economic disruptions during the transition process, necessitating targeted policy interventions and support measures [60,61]. Central to the effectiveness of energy transition policies is the role of government and institutional capacity in steering the transition process. The analysis highlights the importance of proactive policy leadership, regulatory enforcement, and capacity-building efforts at the provincial level to ensure the implementation of national energy and environmental objectives [62]. Moreover, enhancing coordination and collaboration between government agencies, industry stakeholders, and civil society actors is essential for fostering synergy and overcoming administrative silos and bureaucratic hurdles [63,64].

Technological innovation and market dynamics emerge as critical drivers shaping the energy transition landscape in China. The rapid advancements in renewable energy technologies, coupled with declining costs and supportive government policies, offer unprecedented opportunities for scaling up clean energy deployment and reducing the reliance on fossil fuels [50,51]. Furthermore, market incentives, such as carbon pricing mechanisms and renewable energy subsidies, play a pivotal role in incentivizing investment in clean energy projects and driving market competitiveness [22]. This Discussion underscores the importance of public awareness and stakeholder engagement in fostering societal support for energy transition policies. Public participation in decision-making processes, access to information, and mechanisms for redress are essential for building trust, legitimacy, and social acceptance of energy projects [45,46]. Moreover, fostering partnerships with local communities, indigenous groups, and grassroots organizations can enhance the effectiveness and sustainability of energy transition initiatives by incorporating local knowledge, values, and preferences into policy design and implementation [56,57].

Finally, this Discussion emphasizes the significance of international cooperation and knowledge exchange in advancing energy transition goals in China. Collaborative initiatives, such as joint research projects, technology transfer programs, and capacity-building partnerships, can facilitate the exchange of best practices, lessons learned, and innovative solutions to common challenges [65,66,67]. Moreover, participation in multilateral forums and global initiatives, such as the Paris Agreement and the United Nations Sustainable Development Goals, underscores China’s commitment to addressing climate change and promoting sustainable development on the global stage [68,69,70].

Implementing energy transition and ESG policies at the regional level faces several barriers, including political resistance, budget constraints, and limited institutional capacity. Political resistance can arise from competing priorities, ideological differences, and lack of political will, especially in regions dependent on traditional energy sectors. Budget constraints hinder the necessary investments in clean energy infrastructure and technology, with many regions facing difficulties in allocating resources due to fiscal challenges. Institutional capacity issues, such as weak governance frameworks, a lack of technical expertise, and poor coordination between various stakeholders, can delay or complicate policy execution. Economic and structural barriers, such as regional dependence on fossil fuels and disparities in resources, further complicate the transition, with less developed areas facing greater difficulties. Social and cultural barriers, including limited public awareness, concerns about equity, and resistance from affected communities, can undermine policy acceptance and implementation. Additionally, global market dynamics, fluctuating energy prices, and international pressures can impact the feasibility of regional policies, adding another layer of complexity. Overcoming these barriers requires strong leadership, adequate financial resources, capacity-building, and coordinated efforts to address equity concerns and ensure the transition is inclusive and effective.

The discussion underscores the complexity and urgency of addressing ESG risks and driving factors in China’s provincial-level energy transition policies. By adopting a comprehensive and integrated approach that takes into account environmental protection, social equity, and good governance principles, China can chart a more sustainable and inclusive path towards energy security, economic prosperity, and environmental resilience. Through concerted efforts to enhance policy coherence, institutional capacity, stakeholder engagement, and international cooperation, China can emerge as a global leader in driving the transition towards a low-carbon and resilient energy future.

5. Conclusions

China’s energy transition is a critical component of global efforts to combat climate change and achieve sustainable development. However, the complex interdependencies within the water–energy–land–food (WELF) nexus present significant environmental, social, and governance (ESG) risks that must be carefully managed. This study highlights the need for a holistic governance framework that incorporates cross-sectoral collaboration, improved policy coordination, and stronger ESG risk management strategies. Sustainable energy policies should not only focus on decarbonization but also ensure responsible resource allocation, equitable distribution of benefits, and long-term environmental resilience. Moreover, advancements in technology, circular economy principles, and innovative financing mechanisms can help mitigate ESG risks while fostering a just and inclusive energy transition.

In conclusion, while this study provides valuable insights into the impact of energy transition policies on regional ESG risks, there are several areas that could be strengthened for greater clarity and depth. Limitations in the data, challenges facing during policy implementation, and the causal mechanisms involved are factors impacting the robustness of the findings. Addressing these factors would not only improve the validity but also increase its practical utility for policymakers and researchers, providing a more comprehensive framework for understanding and addressing the complexities of energy transition and ESG risks.