[go: up one dir, main page]
More Web Proxy on the site http://driver.im/
Next Article in Journal
Source Apportionment and Analysis of Potentially Toxic Element Sources in Agricultural Soils Based on the Positive Matrix Factorization and Geo-Detector Models
Previous Article in Journal
Research on “Identification–Cognition–Perception” of the Pedestrian Spaces Around Subway Stations near Popular Tourist Attractions from the Tourists’ Perspective: A Case Study of Tianjin
You seem to have javascript disabled. Please note that many of the page functionalities won't work as expected without javascript enabled.
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Land
Volume 14
Issue 1
10.3390/land14010147
land-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Facilitating or Hindering? The Impact of Low-Carbon Pilot Policies on Socio-Ecological Resilience in Resource-Based Cities

by
Yanran Peng
,
Zhong Wang
*,
Yunhui Zhang
and
Wei Wang
School of Public Administration, China University of Geosciences, Wuhan 430074, China
*
Author to whom correspondence should be addressed.
Land 2025, 14(1), 147; https://doi.org/10.3390/land14010147
Submission received: 4 December 2024 / Revised: 7 January 2025 / Accepted: 11 January 2025 / Published: 13 January 2025
(This article belongs to the Topic Advances in Environmental Behaviors of Persistent Contaminants: Fate, Distribution, Risk, and Challenges)
Browse Figures
Figure 1
<p>Study area pilot.</p> "> Figure 2
<p>Kernel density of spatial differentiation of the socio-ecological resilience of resource-based cities, 2003–2022. (<b>a</b>) Socio-ecological resilience kernel density; (<b>b</b>) Economic resilience kernel density; (<b>c</b>) Social resilience kernel density; (<b>d</b>) Ecological resilience kernel density.</p> "> Figure 3
<p>Parallel trend test results. The solid line represents the estimated dynamic policy effects over time, with the vertical dashed lines indicating the 99% confidence interval. The horizontal line (zero line) serves as the baseline to assess whether the policy effect differs significantly from zero, while the vertical dashed line on the x-axis marks the policy implementation time, establishing the baseline for the test.</p> "> Figure 4
<p>Placebo test results. The placebo test results are shown as follows: The solid line represents the kernel density estimation of the variable, reflecting the distribution of the estimated values. The dashed line marks the reference significance level (baseline at <span class="html-italic">p</span>-value = 0.10), while the blue dots represent the <span class="html-italic">p</span>-values for each estimated value, illustrating how significance levels change with the estimated values.</p> "> Figure 5
<p>Variable differences before and after propensity score matching. This figure is used to assess whether the PSM effectively balances the distribution of covariates between the treatment and control groups. If the matched points are closer to zero, it indicates that the PSM has effectively improved the balance of covariates. The matched points in the figure are closer to zero, indicating a good matching effect.</p> "> Figure 6
<p>Local Moran’s Index. Using the plotting module in Stata, local Moran’s scatter plots of socio-ecological resilience for resource-based cities in 2003, 2009, 2015, and 2022 were drawn, to examine whether the results are concentrated along a straight line. (<b>a</b>) Localized Moran’s Index, 2003; (<b>b</b>) Localized Moran’s Index 2009; (<b>c</b>) Localized Moran’s Index 2015; (<b>d</b>) Localized Moran’s Index in 2022.</p> "> Figure 7
<p>Distribution of socio-ecological resilience values for two different types of cities. (<b>a</b>) Socio-ecological resilience of HHM; (<b>b</b>) Socio-ecological resilience of MML.</p> "> Figure 8
<p>Policy implications for socio-ecological resilience in different types of pilot cities. Using double machine learning in Python, the impact coefficients of low-carbon policies are assessed. The plotting module is used to generate a chart of policy impact coefficients, with the vertical axis representing the impact coefficients, *, **, and *** correspond to significance levels of 10%, 5%, and 1%, respectively. (<b>a</b>) Impact of <span class="html-italic">LCCP</span> on the <span class="html-italic">SESR</span> of MML-cities; (<b>b</b>) Impact of <span class="html-italic">LCCP</span> on the <span class="html-italic">SESR</span> of HHM-cities.</p> ">
Review Reports Versions Notes

Abstract

:
Low-carbon pilot policies are essential for the green transformation of resource-based cities, helping them mitigate the “carbon curse” and the “resource curse” while promoting sustainable socio-ecological development. Focusing on a panel of 114 resource-based cities in China, spanning from 2003 to 2022, this study employs a range of methodologies, including kernel density estimation, the Difference-in-Differences Model, Spatial Difference-in-Differences, Mediation Analysis, K-means Clustering, and Dual Machine Learning to assess the consequences of low-carbon pilot policies on socio-ecological resilience. The findings indicate that the socio-ecological resilience of the study area has generally improved, though there is noticeable polarization. Low-carbon pilot policies significantly enhance the resilience of resource-based cities by 0.4%, and they exhibit a positive spatial spillover effect of 1.1%. However, the long-term effects of the policies on economic resilience were not significant, and the policies did not have a direct impact on the social resilience of the pilot cities; however, they did promote social resilience in neighboring regions. Finally, the effectiveness of low-carbon pilots varies, with more pronounced benefits in declining and mature resource cities, particularly in those with medium ecological and economic resilience, and low social resilience. Green finance, industrial transformation, and carbon emission efficiency are identified as key strategies for improving socio-ecological resilience. The above findings provide insights for policymakers seeking to foster inclusive, resilient, and sustainable urban development in China.

1. Introduction

The “resource curse” and the “carbon curse” are two significant challenges frequently encountered by resource-based cities. These cities are characterized by industries centered on the extraction and exploitation of natural assets, such as minerals and timber. This results in the depletion of resources and high carbon emissions, both of which pose a significant threat to the sustainability of their socio-ecological systems [1]. These cities represent a critical “bottleneck” in China’s broader efforts to enhance socio-ecological resilience. The industrial development of resource-based cities is largely influenced by resource extraction. The mining sector, in particular, accounts for over 20% of secondary industry. This reliance on mining has led to the growth of industries that are associated with high emissions, significant environmental damage, and substantial consumption of energy. Their rough and expansive development paths have resulted in resource depletion, diminished growth prospects, reduced industrial efficiency, rising unemployment, and environmental degradation, exposing the vulnerabilities of these regions across economic, social, and ecological dimensions. Consequently, these urban systems exhibit limited capacity to withstand external shocks, and their socio-ecological resilience remains inadequate.
Resource-based cities play a central role in China’s strategy for enhancing socio-ecological resilience. Approximately 40% of cities in China rely on natural resources and will continue to be vital hubs for energy and resource supply in the medium to long term [2]. These cities are essential for advancing China’s high-quality economic development. The National Sustainable Development Plan for Resource-Based Cities (2013–2020) [3], issued by the State Council, emphasizes the importance of promoting green, low-carbon, and regenerative development in resource-based cities. Similarly, the European Commission’s “European Green Deal”, released in 2019, and the United States’ Clean Energy Revolution and Environmental Justice Plan, proposed in 2020, aim to establish a positive synergy between resource extraction and low-carbon urban social-ecological development., emphasizing the importance of promoting green, low-carbon and regenerative development. In line with this objective, the State Reform Commission’s Pilot Policy for Low-Carbon Cities [4] includes 36 resource-based cities and sets clear directives for industrial restructuring, low-carbon development pathways, and energy optimization. This policy has become a crucial tool for driving the green transformation of the socio-ecological systems within these cities [5].
Despite their potential, resource-based cities represent the most challenging aspect of China’s efforts to bolster socio-ecological resilience. The Low Carbon Pilot Policy (hereinafter referred to as LCCP) is a systematic project that involves many changes in a city’s energy structure, transformation of traditional industries, and other changes that affect the external environment of resource-intensive industries. These changes are expected to influence the production, consumption, and distribution patterns across industries and along the industrial veins of the city, with ripple effects extending to the broader socio-ecological systems and neighboring regions. On the one hand, the LCCP aims to encourage cleaner production practices, transform development models, optimize economic structures, and drive new growth trajectories [6], while simultaneously creating green employment opportunities [7] that enhance both socio-ecological resilience. On the other hand, resource-based cities face huge exit costs for low-carbon development due to their energy dependence and carbon-intensive development paths [8,9]. The low-carbon transition imposes constraints on their traditional resource-intensive industries. In contrast, emerging industries lack the stamina to develop, exacerbating the risks of industrial imbalances and insufficient economic growth, which may lead to substantial unemployment and undermine socio-ecological resilience in the short term.
The current literature has not thoroughly investigated whether low-carbon pilot policies positively impact the socio-ecological resilience of resource-based cities, particularly in terms of their geographical spillover consequences. Consequently, the following empirical questions are the focus of this study: Does low-carbon pilot policy improve resource-based cities’ socio-ecological resilience? Do spatial spillover effects exist? How do the effects differ in the short term versus the long term? What mechanisms drive these impacts?
This study employs panel data from 114 resource-dependent cities in China spanning from 2003 to 2022 and is grounded in resilience theory. It synthesizes various scholarly perspectives and multiple dimensions to develop an evaluation framework for the social-ecological resilience of these cities. Kernel density estimation is utilized to analyze the trends in resilience dynamics. Moreover, the Low-Carbon Pilot Policy is treated as a quasi-natural experiment. The study assesses the impact of the LCCP on the social-ecological resilience of resource-dependent cities using advanced econometric methods, including Difference-in-Differences (DID), Spatial Difference-in-Differences (SDID), K-means clustering, and Double Machine Learning (DML). The analysis considers both overall effects and multidimensional effects, local and spatial effects, as well as the long-term and short-term consequences, encompassing both direct and indirect effects. Additionally, we examine the robustness, heterogeneity, and underlying mechanisms of these impacts.
This study’s innovation and contributions are primarily reflected in three key aspects. First, this study is the first to examine the effects of low-carbon pilot projects on the socio-ecological resilience of resource-dependent cities, encompassing both the overall and dimensional impacts. The study takes into account the unique nature of low-carbon transitions in resource-based cities, as well as the multidimensional and bidirectional effects that low-carbon policies may generate, offering a more comprehensive analysis of policy effects. Secondly, this study introduces an innovative spatial dynamic model. This model can assess both the short- and long-term temporal impacts of the policy, as well as its local and spatial effects. This approach enables a more precise identification of the dynamic effects of the policy. Thirdly, this study can provide a reference for cities with high-carbon development types worldwide to address climate change and promote low-carbon urban social-ecological development and provides valuable insights into the governance of various resilience dimensions of cities in the context of low-carbon development. It also conducts a regional heterogeneity analysis of policy effects, offering relevant guidance for understanding regional characteristics, highlighting individual differences, and providing a reference for the effective implementation of carbon reduction strategies.

2. Literature Review and Research Hypotheses

2.1. Literature Review

Social-ecological resilience is widely defined by scholars as the capacity of a system to withstand external shocks, adapt to changes, and recover [10], emphasizing continuous proactive adaptation and adjustment to achieve sustainable development across multiple dimensions, marking the third stage of urban resilience development. The stronger the ability of a social-ecological system to withstand disturbances and risks, the more resilient it becomes [11,12]. Scholars adopt various perspectives and methodologies when studying social-ecological resilience. Theoretical analysis focuses on analytical frameworks, assessment indicators, and models of resilience in social-ecological systems. For example, Bruneau developed the “TOSE” resilience identification framework for regions [13], which includes four dimensions: technology, organization, society, and economy. Oliveira considering system feedback, non-linearity, and path dependency, established a prototype model of social-ecological system resilience [14]. Talubo selected indicators from social, economic, and ecological dimensions to assess the disaster resistance capacity of social-ecological systems [15].
From a spatial evolutionary perspective, scholars have focused on urban areas [16], counties [17,18], and specific regions [19]. Primarily utilizing composite indices [20], they have constructed region-specific social-ecological resilience assessment frameworks to measure resilience and reveal the evolution patterns and influencing mechanisms of social-ecological systems. Research indicates that the dominant factors influencing the evolution of resilience in social-ecological systems vary significantly across different regional types. Given the varying stages of urban development, diverse strategies can be adopted to enhance the resilience capacity of urban areas. These strategies include prevention and monitoring, emergency and buffering, recovery and reconstruction, as well as innovation and transformation.
Given the ongoing global warming, climate change has become a major global concern, posing a serious threat to urban socio-ecological systems. At present, research on urban resilience primarily emphasizes low-carbon development and the impacts of climate change. Low-carbon city pilot projects represent some of the largest decarbonization initiatives both within China and globally [21]. Currently, several notable models of low-carbon city development include Denmark’s low-carbon communities, the UK’s Urban Climate Change Action, Sweden’s Sustainable Action Plan, Japan’s Low-Carbon Society Action Plan, the US’s Low-Carbon City Action Plan, and China’s Low-Carbon Pilot Programs, among others. Most of these models focus on transforming traditional high-carbon industries. These low-carbon development models have yielded significant successes in reducing urban carbon emissions while balancing energy consumption, carbon emissions, and economic growth [22]. However, scholars such as Luis et al. argue that low-carbon transformation should be introduced gradually, allowing time for communities to adapt to the dynamics, relationships, and uncertainties inherent in economic systems [23].
The low-carbon pilot programs in China have had a comprehensive impact on multiple dimensions of urban socio-ecological resilience. In terms of economic resilience, the LCCP has optimized energy infrastructure, adjusted regional industrial structures [24,25,26], improved resource allocation efficiency, and promoted policy innovation and technological progress [27]. These measures have provided the industrial foundation and technological support for economic resilience development. Regarding ecological resilience, researchers suggest that low-carbon clean production can significantly reduce carbon emissions [28], improve carbon emission efficiency [27], and enhance the ecological environment [29]. In terms of social resilience, scholars argue that low-carbon city pilot programs promote employment growth [30], improve residents’ quality of life [31], and positively impact population dynamics [32]. The LCCP has facilitated low-carbon transformation and economic diversification, creating employment opportunities, alleviating labor surplus, and effectively reducing employment risks associated with job transitions. Therefore, the LCCP has made significant progress in enhancing the social-ecological resilience of urban areas.
As a specific type of city, resource-based cities face numerous challenges and pressures within the context of low-carbon development and resilient urban construction [2]. Existing research mainly focuses on measuring different dimensions of resilience in resource-based cities and identifying the influencing factors. The key dimensions of this research include economic resilience [33,34], ecological resilience [35,36], and overall resilience [37]. As for the influencing factors, scholars argue that industrial structure [38], population shrinkage [39], urbanization [40], carbon emissions [2], and infrastructure development [35] all affect the social-ecological resilience of resource-based cities. For instance, Wang found a positive correlation between environmental management strategies and economic resilience [33], while Zhang suggest that land use efficiency influences ecological resilience through the regulatory role of environmental legislation [41].
A review of the existing literature reveals that scholars generally agree that the resilience of resource-based cities has improved [42] and that low-carbon development positively influences the resilience of these cities. However, low-carbon development is often regarded as an intermediate variable; there remains a gap in research examining the direct relationship between low-carbon development and resilience in resource-based cities. Moreover, existing studies primarily focus on the impact of low-carbon clean production on a single dimension of resilience, with limited exploration of its multifaceted and long-term effects on the socio-ecological resilience of resource-based cities. Finally, there is a notable lack of evaluations regarding the effects of similar low-carbon city initiatives in international research. This study aims to provide a reference framework and research methodology for evaluating policy effectiveness.

2.2. Research Hypotheses

Research has demonstrated that low-carbon development can greatly improve cities’ resilience [43]. As a prominent low-carbon governance practice, the Low-Carbon City Plan (LCCP) aids resource-based cities in enhancing their socio-ecological resilience through various mechanisms. First, resource-based cities are typically reliant on a singular resource extraction industry, which renders them susceptible to economic fluctuations. The LCCP facilitates the establishment of carbon trading markets, promotes industrial restructuring, and significantly supports the development of low-carbon industries. This fosters economic diversification, thereby strengthening the stability and adaptive capacity of the city’s economic system. Additionally, low-carbon industries generate employment opportunities, mitigate social challenges arising from industrial decline—such as unemployment and instability—and support societal adaptation to low-carbon transition reforms [44]. Second, resource-dependent cities frequently confront the dual challenges of over-extraction of resources and ecological degradation. Low-carbon policies emphasize sustainable resource utilization and environmental protection, promote the widespread adoption of reducing carbon emissions through the use of low-carbon technologies, and foster ecosystem restoration through improvements in energy structure and carbon efficiency [5,45]. The socio-ecological resilience of cities is a multidimensional, interconnected system, where the social, economic, and ecological spheres interact. By implementing reforms across multiple dimensions, low-carbon pilots can foster coordinated development, thereby enhancing the overall socio-ecological resilience of resource-dependent cities.
However, LCCP also poses challenges: it inhibits the development of resource extraction and related industries, potentially slowing economic growth. This decline in economic activity could adversely affect social welfare [46]. Therefore, a comprehensive evaluation of the integrated impacts of LCCP on the socio-ecological resilience of resource-dependent cities is essential.
Finally, the study of spatial spillovers from LCCP is especially pertinent within the Chinese policy context, where local governments exhibit interactive behaviors, such as learning, imitation, and policy competition, which can lead to spillovers through political and economic linkages. There is the possibility of a “free-rider” effect, where factors of production—such as technology, capital, and knowledge—diffuse from more mature pilot cities to neighboring underdeveloped regions, fostering technological progress and industrial upgrading [47]. Non-pilot cities can improve their socio-ecological resilience by adopting and imitating the innovative technologies and development strategies of core cities. Moreover, the inherent competition among local governments may drive non-pilot cities to replicate the policy initiatives of the LCCP in order to realize both economic and environmental gains. The following Hypothesis 1 is proposed in light of the previous analysis:
Hypothesis 1. 
The implementation of the LCCP has substantially strengthened the socio-ecological resilience of pilot resource-dependent cities, with its effects proving stable over both the short and long term. Furthermore, it exerts a significant influence on neighboring cities.
According to existing theories, low-carbon pilots improve socio-ecological resilience through the following mechanisms.
First, Green Finance (hereinafter referred to as GF), an essential tool for achieving global sustainable development, provides long-term and stable financial support for pilot policies, including a range of financial products and plans, such as green insurance, green funds, and green credit [48]. GF directs capital towards projects focused on environmental protection and low-carbon industries, mitigates the financial challenges associated with low-carbon transitions, and spreads environmental risks. This approach establishes a mutually beneficial strategy that promotes both economic growth and environmental sustainability. Through the guidance and incentives of green finance, enterprises and individuals could gain greater access to resources for investing in the study, growth, and use of low-carbon technologies [49], thereby accelerating the green transformation of industries. Due to the existence of market mechanisms, GF has a positive knock-on effect on neighboring communities in addition to improving the socio-ecological resilience of pilot towns via capital flows and technology diffusion. These flows of capital fuel the expansion of the environmental protection industry, while technological innovations promote the upgrading of green technologies across the region, boosting regional socio-ecological resilience to climate change.
Second, industrial restructuring (hereinafter referred to as IS) is not only the key path of low-carbon pilot policies, but also an essential strategy for improving resource efficiency and alleviating ecological strain [50]. By phasing out highly polluting and energy-intensive industries, supporting clean energy sectors, and promoting the adoption of green technologies and circular economy models, pilot cities optimize resource utilization, reduce environmental pollution, and ease pressure on ecosystems [51]. Industrial restructuring is a critical method for reducing carbon emissions; through policy guidance, cities have increased investments in clean energy, green technology, and sustainable industries, while gradually eliminating outdated, highly polluting production capacities. This transformation has opened up opportunities for green growth in the urban economy, fostering economic diversification and resilience, and significantly enhancing societal capacity to withstand external shocks, such as climate change and resource depletion. Additionally, industry interconnectivity, knowledge-sharing, and resource flows have allowed the results of industrial upgrades in pilot cities to positively impact surrounding areas. These areas can emulate the pilot cities’ experiences, optimizing their own industrial structures and applying green technologies, thereby boosting their overall socio-ecological resilience.
Finally, improvements in carbon efficiency (hereinafter referred to as CE) are a key outcome of low-carbon transitions [5]. By limiting cities’ energy dependence and resource extraction, pilot policies have reduced carbon emissions and directly improved environmental quality and ecosystem stability. Improvements in carbon efficiency (defined as a reduction in carbon emissions per unit of output) have created a mutually beneficial situation for economic growth and environmental sustainability. By promoting the use of clean energy and low-carbon technologies, these policies have effectively reduced energy consumption and carbon emissions in the process of economic output, thus promoting the development of green production methods. This not only helps to alleviate the double dilemma of resource depletion and economic development faced by resource-based cities, but also maximizes the use of limited resources, improves economic output, and achieves a win-win situation for economic growth and environmental protection. In addition, the popularization of low-carbon technologies and green production methods is not limited to pilot cities, and their spillover effects are gradually radiating to surrounding areas, promoting the low-carbon transformation of the entire region. These measures not only improve the socio-ecological resilience of resource-based cities, but also enhance their adaptability to future environmental and resource challenges, helping to alleviate the current fragile socio-ecological system:
Hypothesis 2a. 
The LCCP enhances the socio-ecological resilience of pilot cities and neighboring areas through green finance.
Hypothesis 2b. 
The LCCP improves the socio-ecological resilience of the pilot cities and neighboring areas by promoting industrial structural upgrading.
Hypothesis 2c. 
The LCCP enhances the socio-ecological resilience of the pilot cities and neighboring areas by improving carbon emission efficiency.

3. Research Design

To assess the impact of the LCCP on socio-ecological resilience, we constructed a panel dataset covering 114 resource-based cities from 2003 to 2022, comprising 2280 observations. In accordance with the principles of comprehensiveness, reliability, traceability, and comparability, data were obtained from authoritative sources, including the China Statistical Yearbook, China Urban Statistical Yearbook, China Urban Construction Statistical Yearbook, and the National Economic, as well as the China Carbon Accounting Database and Social Development Bulletin published by each city. To ensure continuity, missing data points were interpolated using the moving average method.

3.1. Research Methodology

3.1.1. Difference-in-Differences Model (DID)

We use the LCCP as a quasi-natural experiment to develop a Difference-in-Differences (DID) model, which evaluates the net effect of the LCCP on the socio-ecological resilience (hereafter referred to as SESR) of resource-dependent cities. Cities participating in the LCCP are referred to as the treatment group, while other cities serve as the control group, as shown in Equation (1):
S E S R i t = α 0 + α 1 L C C P i t + β ∑ X i t + λ i + ν t + ε i t
In this model, i and t represent the city and year, respectively, while SESR denotes socio-ecological resilience. The term α0 represents the intercept, and L C C P serves as the core explanatory variable, with the coefficient α1 indicating its marginal effect on SESR. If α1 > 0, this implies that L C C P enhances the S E S R of the city, and vice versa. Xit denotes a set of control variables, and β represents the coefficient for the Xit estimation parameter, λi accounts for city-specific fixed effects, νt captures time-specific fixed effects, and εit is the random error term.
In order to examine the LCCP’s effects on the SESR of resource-dependent cities, we also consider potential time-dependent effects, as well as issues of endogeneity and policy lag. Furthermore, through spatial dispersion and polarization effects, the LCCP is anticipated to have an impact on nearby non-pilot cities in addition to the pilot cities. To take this into consideration, and using the process of Wang et al. [52], this paper extends the DID model and constructs a Spatial-Difference-in-Differences model based on the equation as shown in (2).
S E S R i t = β 0 + β 1 S E S R i , t - 1 + ρ 1 ∑ i = 1 n w i j S E S R j t + β 2 L C C P i t + ρ 2 ∑ i = 1 n w i j L C C P j t + δ ∑ X i t + λ ∑ i = 1 n w i j X j t + λ i + v t + ε i t
Considering that the spatial correlation between LCCP and SESR is shaped by both geographical and economic factors, we constructed an economic geography matrix, denoted as W, consisting of spatial latitude-longitude distances and per capita GDP. In the equation SESRi,t−1 represents the lagged SESR, which is used to control for and examine the time lag in SESR changes; LCCPit denotes the matrix of low-carbon pilot policy components; ρ1, ρ2, and λ are spatial lag terms; and β0, β1, β2, and δ are the coefficients to be estimated.

3.1.2. Moderated Effects Model

We employ mediation effect analysis to explore potential mechanisms, introducing model (3), where M represents the mediator, with the definition and statistical description of the mediator provided in Section 3.2. The direct and spillover effects of LCCP on the mediator can be evaluated by examining φ1 and φ2 as outlined below.
M it = α 0 + ∑ j W ij M it + φ 1 L C C P it + φ 2 ∑ j W ij L C C P it + β control it + γ ∑ j W ij control it + μ i + ν t + ε it
In line with the approach of [53], we introduce mediators into the equation, leading to Equation (4).
S E S R it = α 0 + ρ ∑ j W ij S E S R it + η 1 L C C P it + η 2 ∑ j W ij L C C P it + η 3 M i t + η 4 ∑ j W i j M i t + B c o n t r o l s it + γ ∑ j W ij control it + μ i + ν t + ε it
According to relevant studies, the local impact of the LCCP through mediation effects can be calculated as φ1 × η3, while its effect on neighboring cities can be determined as φ2 × η3 [54]. Thus, the mechanism through which the LCCP affects neighboring SESR can be understood by integrating the indirect effect of the LCCP on the mediator with its local effects.

3.1.3. K-Means Cluster Analysis

The fundamental idea behind K-means cluster analysis is its capacity to classify data samples by comparing their distances; the closer two data points are to one another, the more similar they are. The algorithm starts by randomly selecting initial centroids, after which it calculates the distances between these centroids and the other data points. Equations (5) and (6) define the K-means algorithm’s objective function.
J = ∑ j k ∑ x i ∈ w i ∥ x i − c j ∥ 2
c j = 1 n ∑ x i ∈ w i x i
In this context, J denotes the objective function of K-means, reflecting the distance relationship between the data points in the dataset and the centroids. ωi represents the sample dataset, xi refers to the i-th sample in the dataset, and cj denotes the centroid of the j-th cluster.

3.1.4. Double Machine Learning (DML)

We utilize Double Machine Learning (DML) for causal analysis, a technique that is well suited for handling high-dimensional datasets and mitigating bias. DML effectively accounts for non-linear relationships among variables, thus reducing the biases that may arise from such relationships. The following Equations (7) and (8) formalize the DML model.
S E S R = θ x T + g X + ε   w h e r e   E ε | T , X = 0
T = f X + τ   where   E τ | X = 0
In these equations, SESR serves as the primary indicator of the experimental effect, reflecting the SESR of resource-based cities. T is the treatment indicator, a binary variable (0/1) that signifies whether a sample belongs to the experimental (LCCP) group or the control group, while T⊥X denotes a randomized A/B experiment. θ(X) denotes the covariate. X encompasses characteristics that are not influenced by the experimental intervention, typically represented as a high-dimensional vector.

3.2. Variable Selection and Measurement

3.2.1. Explanatory Variable

Low-carbon city pilot data were manually collected from the National Development and Reform Commission (NDRC). In 2010, 2012, and 2017, the Chinese government launched three cohorts of LCCP cities. The selection of 114 resource-based cities as the study sample was based on the list provided in the “National Sustainable Development Plan for Resource-Based Cities (2013–2020)”, announced in 2013 by the State Council, considering the availability and comparability of urban data. By cross-referencing this list with the LCCP, the final sample comprises 36 LCCP cities as the experimental group and the remaining 78 cities as the control group, as shown in Figure 1. A policy variable, LCCP, was created by multiplying a set of dummy variables (representing LCCP and non-LCCP regions) with time dummy variables (before and after policy implementation).

3.2.2. Dependent Variable

Socio-ecological resilience refers to the capacity of urban systems to endure crises, disasters, and disruptive shocks, as well as their ability to recover from them [55]. Guided by principles such as analytical reliability, measurability, national representativeness, relevance to urban resilience, and interdependencies, we developed an evaluation index system consisting of three dimensions: economic prosperity, social welfare, and environmental quality, as shown in Table 1. The SESR was calculated using the entropy-weighted TOPSIS method. Data on economic resilience, social resilience, and ecological resilience are derived from the Economic Development, Social Services and Infrastructure, and Energy and Environment categories of the China Urban Statistical Yearbook, respectively.
  • Economic Prosperity: Economic resilience allows urban economies to operate efficiently and adapt effectively to external shocks, thereby ensuring sustained economic growth through optimal resource allocation, industrial advancement, and structural adaptations [38]. This dimension is represented by five indicators: economic strength, economic vitality, industrial structure, investment efficiency, and green development.
  • Social Welfare: Social resilience is centered on human well-being, ensuring that urban residents can live harmoniously, participate in collective activities, and access educational and cultural resources, highlighting the level of quality of urban services and social infrastructure [56]. This dimension is represented by five indicators: fiscal capacity, employment pressure, social security, education level, and public services. Furthermore, resources and facilities should be measured by their accessibility [57], and consequently, per capita values or ratios are employed for these indicators.
  • Environmental quality: A robust ecosystem is the foundation for urban resilience against natural catastrophes and environmental pollution. A city’s ecological resilience is reflected in its urban environment and protection measures [58]. This dimension is represented by five indicators: ecological load, green services, greening level, pollution control, and environmental management.
To address potential estimation biases arising from significant data discrepancies, we applied a logarithmic transformation to per capita GDP, per capita disposable income, and per capita social security expenditure. Additionally, other data were standardized using tail-trimming techniques.
Table 1. Construction of socio-ecological resilience indicators.
Table 1. Construction of socio-ecological resilience indicators.
DimensionMeaning of IndicatorsIndicators (Unit)WeightsAttributeIndicator-Related Studies
Economic
Prosperity		Economic StrengthGDP per capita (CNY)5.193+(1) Economic development, income level (Han et al.) [59];
(2) Productivity and standard of living, industrial structure, financial resources of the population, openness of the economic system, productivity, level of technological innovation (Shi et al.) [60];
(3) Resistance and recovery ability, adaptation and regulation ability, innovation and transformation ability (Li et al.) [61];
(4) Growth rate of GDP (Feng et al.) [62]
Economic VitalityUrban residents’ disposable
income (CNY)		8.193+
Industrial StructureTertiary industry output
in GDP (%)		4.511+
Investment EfficiencyFixed asset investment
(% of GDP)		18.043−
Green developmentEnergy consumption per
unit of GDP (ton)		0.161−
Social
Welfare		Fiscal capacityGovernment financial
self-sufficiency rate (%)		8.992+(1) Education development, health care level (Han et al.) [59];
(2) Employment situation, transport construction, social capacity to provide education, health care and cultural services (Shi et al.) [60];
(3) Education, employment, health, and infrastructure (Xiong et al.) [63].
(4) Aging rate, proportion of the population with tertiary education or higher, employment in the tertiary sector, number of higher education institutions, research and development personnel, and the count of registered unemployed individuals in urban areas (Xu et al.) [64].
Employment pressureUrban registered
unemployment rate (%)		3.649−
Social securityNumber of health beds
per 104 inhabitants (unit)		1.37+
Education levelNumber of full-time university teachers per 104
inhabitants (unit)		25.773+
Public ServicesSocial security expenditure
per capita (CNY)		3.732+
Clean
Environment		Ecological loadSulphur dioxide emissions
per unit of GDP (ton)		0.251−(1) Environmentally friendly (Han et al) [59];
(2) Greening emission reduction measures, emission levels, pollution control levels (Shi et al.) [60];
(3) Resistance and resilience (Shi et al.) [65];
(4) Resistance, recovery, adaptability (Ning et al.) [40];
Green ServicesGreen park area per capita (HA)3.182+
Greening LevelGreen cover of urban
built-up areas (%)		4.12+
Pollution ProtectionCentralized treatment rate of wastewater treatment plants (%)8.558+
Environmental GovernanceHarmless treatment rate of
domestic rubbish (%)		4.272+
Note: “+” for positive impacts, “−” for negative impacts.

3.2.3. Control Variables

Several other factors also affect the effectiveness of the LCCP. Therefore, in addition to examining its impact on urban socio-ecological resilience, controlling for the effects of other variables is essential. In this research, the following control variables were included in the model: urbanization level (Urb), per capita wage level (Awa), population density (Pud), science and technology expenditures (Pes), share of fiscal environmental protection expenditures (Gsu), and per capita road area (Rap). Data from the China Urban Statistical Yearbook, China Rural Statistical Yearbook, China Urban Construction Statistical Yearbook, and the National Economic and Social Development Bulletin published by each city were used.

3.2.4. Mechanism Variables

We investigate the mediating mechanisms through which green finance, industrial transformation, and carbon emission efficiency affect the relation between the LCCP and the SESR of resource-based cities.
(1) Green Finance: GF focuses on sectors like environmental protection, energy conservation, and renewable energy. The goal is to reallocate financial resources towards more sustainable and eco-friendly sectors. In China, the scope of green finance includes green credit, securities, insurance, and investment. However, due to limitations in data availability at the prefecture-level city scale, this study uses green credit as a proxy for green finance. Specifically, green credit is defined as the proportion of total interest expenditures allocated to industries with lower energy consumption, excluding the six most energy-intensive and polluting sectors. Data from the China Industrial Statistical Yearbook, China Energy Statistical Yearbook, People’s Bank of China, and other authoritative institutions are used.
(2) Industrial Transformation: IS refers to the structural shift in the economy, notably, the transition from manufacturing-based industries (secondary sector) to service-oriented industries (tertiary sector). In this study, industrial transformation is measured by the ratio of the output value of the tertiary industry to that of the secondary industry. This ratio reflects the trend towards a service-driven economic structure and indicates the extent of the shift in the industrial structure. Data from the China Industrial Statistical Yearbook are used.
(3) Carbon Emission Efficiency: CE is calculated by treating energy consumption as an input variable, while carbon emissions are considered as undesirable outputs. Input variables include labor, capital, and energy consumption. The expected output is GDP; the undesirable output is carbon emissions. The super-efficiency SBM model estimates carbon emission efficiency. Data from the China Carbon Accounting Database are used.
Descriptive statistical data for the dataset are provided in Table 2.

4. Empirical Analysis and Discussion

4.1. Dynamic Evolution Analysis of the Socio-Ecological Resilience of Resource-Based Cities

We employ kernel density analysis to comprehensively examine the socio-ecological resilience of resource-based cities in China. The analysis focuses on the distribution of gravity centers, evolution trends, and the degree of differentiation across the three components—social, economic, and ecological resilience—over six time periods: 2003, 2007, 2011, 2015, 2019, and 2022, as illustrated in Figure 2.
The key findings are as follows:
(1) Socio-Ecological Resilience Trend: Firstly, the center of gravity of the kernel density curve in Figure 2a shows a gradual shift to the right, indicating an upward trend in the SESR of resource-based cities—resilience is improving. Second, the kernel density curve’s peak shows an evolutionary pattern of “low-high-low”, which means that the differences in SESR between China’s resource-based cities get smaller at first and then get bigger. Finally, the kernel density curve reveals a notable tailing phenomenon. The right tail extends further to the right each year, with a right sub-peak appearing in 2011. Meanwhile, the left tail gradually shortens, suggesting a growing polarization in the SESR of China’s resource-based cities; this suggests that some cities have surpassed the average level and become leading regions, while others with lower resilience have gradually reached the average level.
(2) Economic Resilience Evolution. In Figure 2b, the economic resilience curve shifts to the right at the fastest rate, indicating relatively rapid improvement. Concurrently, the curve undergoes a transformation, shifting from a high, slender profile to a shortened, wider form with a slender right tail. This indicates that the gap in economic resilience is widening, emphasizing the necessity to address the increasing spatial disparity in economic resilience.
(3) Social Resilience Stabilization. In Figure 2c, the width and height of the social resilience curve stabilizes around 2007, suggesting that the absolute difference in social resilience among resource-based cities has gradually become stable over this period and shows slow growth.
(4) Ecological Resilience Fluctuations. In Figure 2d, the overall temporal trend indicates a reduction in the bandwidth of ecological recovery capacity over time, suggesting that the disparity in ecological resilience between resource-dependent cities is diminishing. However, in 2011, a notable widening of the ecological resilience bandwidth occurred, and the left trailing tail became longer, indicating that the gap between cities has become wider and a portion of the city has become an ecological resilience leader, likely attributable to the initiation of the LCCP in 2010. These pilot cities embarked on low-carbon reforms, which significantly enhanced their ecological resilience. Due to the policy’s lag effect, non-pilot cities experienced no immediate impact, thereby temporarily increasing the gap between resource-dependent cities. As the policy pilot programs expanded and their demonstration effect took hold, the ecological environment of other resource-dependent cities gradually improved, leading to a convergence in resilience levels. Consequently, the regional gap steadily diminished.

4.2. Local Effect

4.2.1. Analysis of DID Results

Table 3 presents the findings of the baseline regression analysis investigating the influence of the LCCP on the SESR of resource-dependent cities. The adjusted R-squared (Adj. R2) improvement indicates that the model fit gradually improves with the incorporation of control variables and bidirectional fixed effects. In Models (1)–(4), the impact of the LCCP on the SESR is statistically significant at the 1% level; the positive coefficient of the LCCP variable (>0) indicates a positive association, thereby substantiating Hypothesis 1: LCCP initiatives bolster the SESR of resource-dependent cities.
In order to mitigate the potential bias of policy impacts resulting from the exclusion of city and time fixed effects, this study designates Model (4) as the reference model for subsequent analysis. The subsequent regression analysis focuses on the three components of SESR—social resilience, economic resilience, and ecological resilience—represented in Models (5)–(7), respectively. The results indicate that the key explanatory variables for social and economic resilience are statistically significant at the 5% threshold, while ecological resilience reaches a significance level of 1%. These findings demonstrate that the LCCP exerts a substantial positive impact across all three dimensions of resilience: social, ecological, and economic.

4.2.2. Robustness Test

Parallel Trend Test

The validity of the DID regression results relies on the assumption that the pre-policy trends in socio-ecological resilience for both the experimental and control groups follow parallel trajectories. Suppose the socio-ecological resilience trends of the experimental and control groups differ before implementing the LCCP; in that case, it suggests that changes in urban resilience are driven by pre-existing time trends rather than by the LCCP itself. The findings presented in Figure 3 demonstrate that the coefficients estimated for the LCCP are only statistically notable after its implementation and fall within the 99% confidence interval. This indicates that before the LCCP implementation, there was no remarkable distinction in the socio-ecological resilience levels between the experimental and control groups, thus fulfilling the parallel trends assumption; the study sample passed the parallel trend test.

Placebo Test

In real-world scenarios, unobservable factors could affect improvements in socio-ecological resilience. To account for this potential interference, a dummy interaction term was constructed for the placebo test, assigning a random selection of cities to a hypothetical treatment group and the remaining cities to the control group. This procedure was repeated in 1000 iterations, generating 1000 virtual estimated coefficients. Figure 4 shows the regression coefficients and their distributions. They tend to cluster around zero, resulting in a bell-shaped curve. By calculating the 99th percentile of the original coefficient (0.007) relative to these virtual coefficients, the test results demonstrate statistical significance at the 1% level. This lends further support to the placebo test and suggests that the LCCP has a substantial impact on the socio-ecological resilience of resource-based cities not due to other unobserved factors. The study’s results are therefore robust.

Exclusion of Other Policies

When implementing the LCCP, other concurrent policies may also influence urban low-carbon transitions. Failing to account for the effects of these policies could lead to biased baseline estimations. In line with the approach of Pan et al. [66], this study controls for three similar environmental policies for the same period shocks: The New Energy Demonstration Pilot (NEDP), The Green Finance Reform and Innovation Pilot Zones (GFRI), and The Key Air Pollution Control Regions (KAPC). The results presented in Models (8)–(10) of Table 4 demonstrate that, after controlling for the influence of these three concurrent policies and accounting for both cities fixed effects and annual city fixed effects, the coefficient for the LCCP variable remains significantly positive. This finding suggests that the LCCP continues to significantly enhance the SESR of resource-based cities, with the results considered robust.

Test of the Propensity Score Matching Method (PSM-DID)

LCCP are selected through local declarations and NDRC approval. The pilot program cities were not randomly selected. The selection mechanism may introduce bias and endogeneity concerns. We used PSM-DID to account for potential bias. To ensure comparability between the experimental and control groups prior to the LCCP, control variables were utilized as matching criteria. A logit model estimated the probability of each resource-based city being selected as a low-carbon pilot. A 1:3 k-nearest neighbor matching process without replacement was performed, with unmatched cities excluded. The benchmark model was re-estimated using the matched sample. Figure 5 shows that the PSM-DID analysis reduced the discrepancy between the experimental and control groups, validating the matching procedure. The DID model in Table 4 shows that the LCCP positively impacts the socio-ecological resilience of resource-based cities, even after addressing potential selection biases (Model (11)).

Dual Machine Learning (DML)

As multicollinearity can distort the results and introduce potential bias in the DID analysis, we use the DML to validate the relationship between the LCCP and SESR. The DML approach effectively reduces estimation bias in causal inference, thereby improving the accuracy and robustness of estimated policy effects. The regression results from the Model (12) in Table 4, display that the LCCP contributes significantly to improving the SESR of resource-based cities. This finding reinforces the conclusions from the benchmark regression analysis.

4.3. Spatial Effect

4.3.1. Spatial Correlation Test

Global Moran’s I Analysis

A fundamental prerequisite for applying spatial econometric models is the spatial correlation of variables; therefore, it is crucial to examine the spatial correlation characteristics of the SESR. To assess this, we employ Moran’s I statistic to test for spatial correlation in SESR. Table 5 presents the global Moran’s I index for the SESR of resource-dependent cities in China. The results indicate that all indices passed the significance tests, confirming the presence of spatial correlation in the observed SESR levels across the selected cities.

Local Moran’s I Analysis

To further explore this, we calculate the local Moran’s I index for SESR, using cross-sectional data from 2003, 2009, 2015, and 2022 for local spatial correlation analysis (Figure 6). The findings reveal a positive slope in the local Moran’s I index, suggesting significant spatial interdependence. Additionally, the SESR of most cities is positioned within quadrants I and III, indicating the presence of “high-high” and “low-low” spatial clustering patterns. This confirms a positive spatial correlation in SESR across cities, thereby justifying the use of spatial econometric models.

4.3.2. Spatial Difference-in-Differences (SDID) Model Test and Regression Results

SDID Model Test

The spatial spillover effect is quantified by the spatial static model, which reflects a global effect. While the global model is useful for identifying overall patterns, it can sometimes lead to biases when interpreting the marginal effects of explanatory variables. Specifically, it does not account for temporal lag effects, which are crucial for assessing the full scope of policy impacts. To address this limitation, we employ the spatial dynamic model (SDM), which includes both time-lagged variables of the explained variable and spatial lag terms. This model differentiates between short- and long-term policy effects, providing a more nuanced understanding of spatial relationships and their evolution over time.
To identify the best spatial panel model, we tested whether to use SAR or SEM. If both models are unsuitable, the Spatial Dynamic Model (SDM) should be used. The SDM was the most suitable model based on the LM test results. The LR and Wald tests were used to see if the SDM could be reduced to an SAR or SEM. The results, in combination with the Hausman Test, show that a spatial dynamic Durbin model with two-way fixed effects is the best for the analysis. Table 6 shows the test results, which confirm the selection of the spatial dynamic Durbin model for the evaluation of policy effects.

SDID Model Results

Table 7 presents the spatial econometric estimation results for the impact of the LCCP on SESR. The following key findings emerge from the analysis:
(1) Positive impact of the LCCP: Following the incorporation of control variables and two-way fixed effects, the main effect coefficient for the LCCP is 0.004, which is statistically significant. This suggests that the implementation of the LCCP in resource-based cities has a positive impact on SESR. Moreover, the short-term and long-term direct effect coefficients are 0.004 and 0.012, respectively, and both exceed the significance threshold. These results demonstrate that the short-term and long-term effects align with the main effect. Furthermore, the spatial lag term (W × CCP) has a coefficient of 0.011, which is also statistically significant, indicating that the LCCP exhibits a positive spatial spillover effect on SESR. The short-term and long-term indirect effect coefficients for this spatial spillover are 0.016 and 0.030, respectively, both of which are statistically significant. The above research results show that the LCCP has a long-term positive impact on SESR in the study area and has a positive spatial spillover effect, thereby corroborating Hypothesis 1.
(2) Social resilience effect: The main long-term direct impact coefficient of social resilience is not statistically significant, indicating that the policy has no obvious long-term impact on the social resilience of the pilot cities. However, the short-term direct impact coefficient is 0.003, which passes the significance test, indicating that the social resilience of the pilot area has improved in the short term. Although the LCCP has no direct long-term impact on the social resilience of the pilot area, it has a positive spatial spillover effect. The short-term and long-term indirect effect coefficients of W × LCCP are 0.012 and 0.070, respectively, both of which are statistically significant, indicating a positive spillover effect on the social resilience of the labor cities.
This result is due to the fact that in the early stages of the low-carbon pilot, the transformation of the industrial structure of the resource-based city was promoted, creating new low-carbon industrial economies and green jobs, and short-term economic growth and employment increased. However, in the long term, the direct closure of the pillar industry—energy-intensive and polluting industries—in the pilot city led to a slowdown in economic growth, which, in turn, led to a decline in welfare levels. At the same time, the transition to low-carbon development led to changes in the structure of labor demand, which further led to unemployment and a decline in social welfare levels [44]. With regard to implementation of low-carbon pilot policies, these policies have stimulated ‘opportunistic decision-making’ by enterprises. A large number of enterprises have relocated to areas with low environmental regulation in order to make more profits, resulting in the hollowing out of local industries and a lack of growth momentum, thereby offsetting the positive impact of the policy. At the same time, surrounding areas also benefit from the ‘free-rider’ effect. The pilot promotion model helps to spread successful low-carbon experiences to surrounding areas, attract enterprises from pilot cities by improving energy conservation and emission reduction technologies, and maximize the positive spillover effects of policies, thereby minimizing losses.
(3) Economic Resilience Effects: The LCCP demonstrated a marked enhancement in the economic resilience of resource-based urban centers in the immediate term. The principal effect coefficient for the LCCP is 0.003, while that for the spatial lag coefficient (W × LCCP) is 0.006. The short-term direct effect coefficient is 0.003, while the short-term indirect effect coefficient is 0.007, both statistically significant. The results demonstrate that the policy has had a favorable impact on the economic resilience of resource-based cities in the short term, with a positive spatial spillover effect. Nevertheless, the long-term impact on economic resilience is not statistically significant. In the initial stages of policy implementation, the LCCP facilitated the transition of resource-based cities from resource-dependent industries to low-carbon and green industries. This resulted in the stimulation of new economic growth points, enhancement of economic diversity, and improvement in resilience to external shocks. Moreover, in the short term, financial assistance during the transition period facilitated industrial growth and the dissemination of low-carbon technologies, contributing to enhanced economic resilience in surrounding areas. However, due to the reliance on heavy industry and the limitations in technology, capital, and markets, the transformation of the economy in resource-based cities is slow and costly. Moreover, the foundation of the nascent industry is fragile, and the growth rate needs to be improved. Consequently, the long-term impact of the policy on economic resilience is constrained. Furthermore, the decline of traditional resource industries gives rise to structural employment issues that impede long-term enhancements in economic resilience.
(4) Ecological Resilience Effects: The LCCP markedly enhanced pilot cities’ ecological resilience, while exhibiting no discernible spatial spillover effects. The principal effect coefficient for the LCCP is 0.015, and both the short-term direct effect coefficient (0.015) and the long-term direct effect coefficient (0.035) are statistically significant. The findings indicate that the implementation of low-carbon development strategies has resulted in an enhancement of the ecological resilience of the pilot cities. The policy has been effective in reducing regional carbon emissions, enhancing urban environmental management, reducing pollution, and facilitating improvements in the local ecological environment.

5. Further Discussion: Impact Mechanisms and Heterogeneity Analysis

5.1. Mechanism Impact Analysis

The preceding empirical study determined that LCCP has a favorable and statistically significant impact on the socio-ecological resilience of resource-based cities and generates a positive spillover effect on surrounding areas. What mechanisms underlie these effects? To explore the impact mechanism, this paper empirically investigates the theoretical pathways based on the preceding analysis.
Table 8 shows the regression results for Equation (3). As shown in the first column, the LCCP benefits green finance (GF) in the local and neighboring cities, indicating that the LCCP enhances socio-ecological resilience through green finance. The second column demonstrates that the LCCP has a detrimental impact on the pilot cities’ industrial structure (IS). The third column demonstrates that the policy exerts a spillover effect on carbon emission efficiency (CE) in non-pilot areas.
Column (1) of Table 9 demonstrates that green finance significantly improves socio-ecological resilience. When considered alongside the findings presented in Table 8, it becomes evident that the LCCP plays a pivotal role in fostering the expansion of green finance at the local and neighborhood levels. Moreover, when mediating variables are taken into account, the impact of the policy on local and nearby cities’ socio-ecological resilience remains robust. This suggests that green finance partially mediates between LCCP and the socio-ecological resilience of local and surrounding cities—Hypothesis 2a is confirmed.
The second column of results indicates that the industrial structure is also a partial mediator in the LCCP’s promotion of urban socio-ecological resilience in the region—Hypothesis 2b is confirmed. When considered alongside Table 8, the findings show that the LCCP has inhibited the industrial transformation in the pilot cities. This can be ascribed to the impact of environmental regulation on the principal industries of resource-based cities, which compels some enterprises to relocate, thereby precipitating a diminution in local industrial activity. Furthermore, the high costs of industrial transformation hinder short-term progress. However, as a mediating variable, industrial transformation and upgrading are crucial in linking low-carbon pilot policies to improvements in socio-ecological resilience, facilitating the region’s overall resilience enhancement. This result reflects the time effect and indirect effect of policies. The policy effect of the LCCP may be non-linear. Low-carbon pilots may complexly impact socio-ecological resilience at different stages through indirect effects (i.e., the mediating effect of industrial transformation). While the LCCP may initially impose certain economic costs or result in a lag effect due to industrial transformation, it ultimately enhances socio-ecological resilience over the long term. Furthermore, the outcomes of low-carbon policies may require an extended period to demonstrate a favorable impact on socio-ecological resilience. This positive effect is more effectively achieved through industrial transformation.
Finally, the results in the third column highlight that carbon emission efficiency partially mediates socio-ecological resilience in surrounding cities through low-carbon pilot projects—Hypothesis 2c is confirmed. The LCCP has a marked impact on the carbon emission efficiency of nearby areas. When resource-based cities implement low-carbon pilot projects, they may introduce green technologies and management models, which can positively influence non-pilot area’s carbon emission efficiency. In addition, such projects can improve the carbon efficiency of surrounding areas through partial industrial transfer, talent mobility, and the establishment of industrial links.

5.2. Heterogeneity Analysis

5.2.1. Resource-Based Urban Heterogeneity

The findings presented in Table 10 indicate that the LCCP can effectively enhance the socio-ecological resilience of both mature and declining resource-based cities. These policies provide new pathways for low-carbon development in these two types of cities.
Mature resource-based cities: Resource development is in a stable phase, extensive development methods have been alleviated, and the industrial structure is in the process of adjustment and upgrading. These cities are usually economically developed and have relatively abundant innovative resources, which can improve the effectiveness of low-carbon governance policies. In contrast, declining cities face economic downturns as their natural resources are depleted and resource-intensive industries can no longer sustain their social ecosystems. Concurrently, the environmental issues arising from extensive development models are becoming more pressing. In this context, implementing the LCCP can foster technological innovation, support industrial transformation, provide a sustainable development model, and offer opportunities for economic and ecological renewal, thereby enhancing socio-ecological resilience.

5.2.2. Socio-Ecological Resilience Heterogeneity

It is assumed that policies can have different effects in different economic, social, and ecological contexts. To investigate these dynamics, this paper employs 15 tertiary indicators of socio-ecological system resilience across 36 pilot cities as clustering factors. Using the elbow method and silhouette coefficient technique, the optimal clustering value of 2 is identified. These 15 indicators were classified and aggregated into three resilience functions: social resilience, economic resilience, and ecological resilience, as shown in Table 11. The data from the two clusters are analyzed using a dual machine learning approach to gain deeper insights into the distribution of socio-ecological resilience across the pilot resource-based cities (Figure 7). The observed trends in the three resilience types for each cluster align consistently, indicating that the resilience dimensions within each cluster exhibit similar characteristics. This alignment strengthens the reliability of the clustering results, highlighting internal coherence in the resilience levels.
As depicted in Figure 7a,b, the first cluster consists of cities with “high economic resilience, high ecological resilience, and medium social resilience” (HHM). Most of these cities have high economic resilience and ecological resilience, but low social resilience as it relates to the well-being of social life, and this category is represented by 16 pilot cities, such as Chizhou, Nanping, etc. In these cities, future government interventions should prioritize enhancing social resilience by increasing funding and support. The second cluster comprises cities characterized by “medium ecological resilience, medium economic resilience, and low social resilience” (MML). These cities exhibit low overall socio-ecological resilience, with significant potential for improvement in all three resilience dimensions—social, ecological, and economic. The representative cities are mainly Baoji, Puer, and 20 other pilot cities.
We deploy the DML to evaluate the policy’s efficacy. To analyze its impact across various categories of pilot cities, a random forest algorithm was trained with 5-fold cross-validation. The results show a clear temporal trend in the policy’s effect, with the influence on both city types intensifying over time within the sample period.
As illustrated in Figure 8a, LCCP has exerted a more pronounced effect on MML cities, significantly fostering the enhancement of social-ecological resilience in these areas. Over time, its impact has progressively strengthened, exhibiting both positive and statistically significant effects. With a substantial increase in the number of pilot cities, by 2021, all pilot cities had demonstrated positive and significant outcomes. The social-ecological systems are particularly vulnerable to external shocks, and the implementation of the LCCP facilitates their transition to low-carbon economies. This transition enables the development of more robust economic and social structures and reduces carbon emissions, thereby mitigating the adverse effects of external shocks. These findings suggest that cities with low resilience require targeted governmental intervention, and the scope of low-carbon pilot initiatives should continue to expand in such cities.
As shown in Figure 8b, the LCCP initially had a positive and significant impact on high-resilience cities. However, in 2022, as the policy was implemented, some cities had already completed their low-carbon transformations, resulting in a gradual attenuation of the policy’s effects. This suggests that the LCCP should be more closely aligned with the development stages of the cities, potentially incorporating market-oriented and financial mechanisms to guide further low-carbon development.

6. Conclusions and Policy Recommendations

6.1. Conclusions

This study utilizes panel data from 114 resource-based cities spanning from 2003 to 2022, employing kernel density analysis, DID, SDID, as well as mediation and heterogeneity analyses to assess the impact of the LCCP on the socio-ecological resilience of these cities. The main findings are as follows:
(1) The socio-ecological resilience of resource-based cities in China has generally improved. Resilience across social, economic, and ecological dimensions has strengthened, with economic resilience exhibiting polarization, while regional disparities in ecological and social resilience have gradually narrowed. (2) The LCCP significantly enhances the socio-ecological resilience of resource-based cities, with notable spatial spillover effects on non-pilot cities. This supports Hypothesis 1; however, the LCCP has no long-term impact on economic resilience. The program exerts a positive short-term effect on the social resilience of pilot cities and generates a positive long-term spillover effect on surrounding cities. Ultimately, the LCCP has led to a long-term improvement in the ecological resilience of pilot cities. (3) The LCCP strengthens the overall socio-ecological resilience of resource-based cities through the promotion of green finance. Furthermore, the policy fosters the socio-ecological resilience of pilot cities by facilitating industrial restructuring. The LCCP also positively influences the socio-ecological resilience of neighboring areas by improving carbon emission efficiency. Thus, Hypothesis 2 is confirmed. (4) The LCCP is particularly effective in enhancing the socio-ecological recovery capacity of mature and declining resource-based cities. The policy demonstrates a greater impact in regions with weaker socio-ecological recovery capacities.

6.2. Policy Recommendations

Firstly, the development model of low-carbon pilot cities should be widely adopted. Simultaneously, resource-based cities in China could benefit from the successful low-carbon community development models employed in countries like Denmark, particularly in the areas of public housing, renewable energy development, and the promotion of efficient resource utilization.
Secondly, policymakers should prioritize leveraging the externalities associated with low-carbon city initiatives, creating both demonstrative and catalytic effects, while carefully considering their scope, direction, and intensity.
Thirdly, drawing on the experience of the Carbon Trust Foundation in the UK’s low-carbon city planning, policymakers could promote innovations in financial products, such as green loans and green bonds, which would attract both social and priority capital into green projects, thereby providing essential financial leverage for economic transformation.
Moreover, green finance should be directed toward supporting industrial transformation and the advancement of low-carbon technologies, fostering industrial diversification, increasing employment opportunities, and enhancing social welfare.
Finally, with reference to Sweden’s Sustainable Action Plan, China should also categorize cities based on their unique characteristics (such as geographic location, environmental policies, economic development, and industrial structure) to ensure the most cost-effective approach to low-carbon development in each city.
This study has yet to explore the optimal scope of low-carbon pilot cities or the feedback mechanisms of urban resilience in response to low-carbon initiatives. Future research should focus on these critical areas.

Author Contributions

Conceptualization, Y.P. and Z.W.; methodology, Z.W.; software, Y.P.; validation, Y.P., Z.W. and Y.Z.; formal analysis, Y.P.; investigation, Y.P.; resources, Y.P.; data curation, Y.P.; writing—original draft preparation, Y.P.; writing—review and editing, W.W. and Y.Z.; visualization, Y.P. and Z.W.; supervision, Z.W.; project administration, W.W.; funding acquisition, Z.W. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by the National Social Science Fund of China, grant number No. 23BGL248.

Data Availability Statement

The data supporting the reported results can be found in the open-access database of the National Bureau of Statistics: https://www.stats.gov.cn/ (accessed on 12 January 2025) and EPS Data Platform https://www.epsnet.com.cn/index.html#/Index (accessed 12 January 2025).

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Wu, J.; Nie, X.; Wang, H. Cur se to Blessing: The Carbon Emissions Trading System and Resource-Based Cities’ Carbon Mitigation. Energy Policy 2023, 183, 113796. [Google Scholar] [CrossRef]
  2. Liu, E.; Wang, Y.; Chen, W.; Chen, W.; Ning, S. Evaluating the Transformation of China’s Resource-Based Cities: An Integrated Sequential Weight and TOPSIS Approach. Socio-Econ. Plan. Sci. 2021, 77, 101022. [Google Scholar] [CrossRef]
  3. Circular of the State Council on the Issuance of the National Sustainable Development Plan for Resource-Based Cities (2013–2020). Available online: https://www.gov.cn/zhengce/content/2013-12/02/content_4549.htm?isappinstalled=0 (accessed on 10 January 2025).
  4. Development and Reform Commission on the Third Batch of National Low-Carbon Cities Pilot Work_epartmental Government Affairs_China.gov.cn. Available online: https://www.ndrc.gov.cn/xxgk/zcfb/tz/201008/t20100810_964674.html (accessed on 10 January 2025).
  5. Zeng, S.; Jin, G.; Tan, K.; Liu, X. Can Low-Carbon City Construction Reduce Carbon Intensity? mpirical Evidence from Low-Carbon City Pilot Policy in China. J. Environ. Manag. 2023, 332, 117363. [Google Scholar] [CrossRef] [PubMed]
  6. Wang, S.; Xiao, S.; Zhang, Q.; Sun, M. How Can Low-Carbon City Construction Enhance Urban Economic Resilience? A Mechanism Analysis Based on Industrial Agglomeration and Technological Innovation Effects. J. Knowl. Econ. 2024, 1–23. [Google Scholar] [CrossRef]
  7. Yip, C.M. On the Labor Market Consequences of Environmental Taxes. J. Environ. Econ. Manag. 2018, 89, 136–152. [Google Scholar] [CrossRef]
  8. Jiao, W.; Zhang, X.; Li, C.; Guo, J. Sustainable Transition of Mining Cities in China: Literature Review and Policy Analysis. Resour. Policy 2021, 74, 101867. [Google Scholar] [CrossRef]
  9. Li, B.; Dewan, H. Efficiency Differences among China’s Resource-Based Cities and Their Determinants. Resour. Policy 2017, 51, 31–38. [Google Scholar] [CrossRef]
  10. Brown, K. Global Environmental Change I: A Social Turn for Resilience? Prog. Hum. Geogr. 2014, 38, 107–117. [Google Scholar] [CrossRef]
  11. Farley, J.; Voinov, A. Economics, Socio-Ecological Resilience and Ecosystem Services. J. Environ. Manag. 2016, 183, 389–398. [Google Scholar] [CrossRef]
  12. Grafton, R.Q.; Doyen, L.; Bene, C.; Borgomeo, E.; Brooks, K.; Chu, L.; Cumming, G.S.; Dixon, J.; Dovers, S.; Garrick, D.; et al. Realizing Resilience for Decision-Making. Nat. Sustain. 2019, 2, 907–913. [Google Scholar] [CrossRef]
  13. Bruneau, M.; Chang, S.E.; Eguchi, R.T.; Lee, G.C.; O’Rourke, T.D.; Reinhorn, A.M.; Shinozuka, M.; Tierney, K.; Wallace, W.A.; von Winterfeldt, D. A Framework to Quantitatively Assess and Enhance the Seismic Resilience of Communities. Earthq. Spectra 2003, 19, 733–752. [Google Scholar] [CrossRef]
  14. Oliveira, B.M.; Boumans, R.; Fath, B.D.; Othoniel, B.; Liu, W.; Harari, J. Prototype of Social-Ecological System’s Resilience Analysis Using a Dynamic Index. Ecol. Indic. 2022, 141, 109113. [Google Scholar] [CrossRef]
  15. Talubo, J.P.; Malenab, R.A.; Morse, S.; Saroj, D. Practitioners’ Participatory Development of Indicators for Island Community Resilience to Disasters. Sustainability 2022, 14, 4102. [Google Scholar] [CrossRef]
  16. Afriyanie, D.; Julian, M.M.; Riqqi, A.; Akbar, R.; Suroso, D.S.A.; Kustiwan, I. Re-Framing Urban Green Spaces Planning for Flood Protection through Socio-Ecological Resilience in Bandung City, Indonesia. Cities 2020, 101, 102710. [Google Scholar] [CrossRef]
  17. Wang, Q.; Chen, L.; Wang, P.; Wang, B.; Yang, T. Evolution of County Socio-Ecological Systems in Nature Reserves in Western China Over the Past 30 Years. Nat. Resour. Res. 2023, 32, 1809–1822. [Google Scholar] [CrossRef]
  18. Wang, T.; Yang, Z.; Han, F.; Yu, J.; Ma, X.; Han, J. Assessment of Tourism Socio-Ecological System Resilience in Arid Areas: A Case Study of Xinjiang, China. Ecol. Indic. 2024, 159, 111748. [Google Scholar] [CrossRef]
  19. Luo, S.; Qiao, D.; Han, X.; Song, B.; Wan, Z.; Li, H. Does a Logging Ban Policy Increase Socio-Ecological Resilience? A Case Study of Key State-Owned Forest Areas in Northeast China. Sustainability 2024, 16, 8368. [Google Scholar] [CrossRef]
  20. Renaud, F.G.; Birkmann, J.; Damm, M.; Gallopin, G.C. Understanding Multiple Thresholds of Coupled Social-Ecological Systems Exposed to Natural Hazards as External Shocks. Nat. Hazards 2010, 55, 749–763. [Google Scholar] [CrossRef]
  21. Zhang, Z.; Zhu, J.; Yang, L.; Chen, L. Booster or Inhibitor: Diagnosing Effects of Low-Carbon Pilot City Policies on Urban Resilience from the Perspectives of Heterogeneity, Mechanisms and Spillover. J. Clean. Prod. 2024, 474, 143587. [Google Scholar] [CrossRef]
  22. Hukkalainen, M.; Virtanen, M.; Paiho, S.; Airaksinen, M. Energy Planning of Low Carbon Urban Areas—Examples from Finland. Sust. Cities Soc. 2017, 35, 715–728. [Google Scholar] [CrossRef]
  23. Mundaca, L.; Busch, H.; Schwer, S. “Successful” Low-Carbon Energy Transitions at the Community Level? An Energy Justice Perspective. Appl. Energy 2018, 218, 292–303. [Google Scholar] [CrossRef]
  24. Qu, F.; Xu, L.; He, C. Leverage Effect or Crowding out Effect? Evidence from Low-Carbon City Pilot and Energy Technology Innovation in China. Sustain. Cities Soc. 2023, 91, 104423. [Google Scholar] [CrossRef]
  25. Yang, S.; Jahanger, A.; Hossain, M.R. Does China’s Low-Carbon City Pilot Intervention Limit Electricity Consumption? An Analysis of Industrial Energy Efficiency Using Time-Varying DID Model. Energy Econ. 2023, 121, 106636. [Google Scholar] [CrossRef]
  26. Luo, Y.; Liu, Y.; Wang, D.; Han, W. Low-Carbon City Pilot Policy and Enterprise Low-Carbon Innovation—A Quasi-Natural Experiment from China. Econ. Anal. Policy 2024, 83, 204–222. [Google Scholar] [CrossRef]
  27. Li, W.; Zhang, Y.; Xu, J.; Fang, S.; Li, Q.; Gong, W.; Wang, C.; Zhang, R. Evaluation for the Effect of Low-Carbon City Pilot Policy: Evidence from Industry in China. Environ. Sci. Pollut. Res. 2024, 31, 8863–8882. [Google Scholar] [CrossRef]
  28. Zhang, J.; Gao, L.; Wang, W.; Deng, Z.; Zhang, X. The Impact of Low-Carbon City Pilot Policies on Air Quality: Quasi-Natural Experimental Evidence from China. Atmosphere 2022, 13, 1355. [Google Scholar] [CrossRef]
  29. Yin, J.; Guo, J. Ecological Effect Assessment of Low-Carbon City Construction in China. Int. J. Environ. Res. Public Health 2022, 19, 14467. [Google Scholar] [CrossRef]
  30. Fu, L.; Zhao, H.; Ma, F.; Chen, J. Estimating Heterogeneous Effects of China’s Low-Carbon Pilot City Policy on Urban Employment. J. Clean. Prod. 2024, 434, 139882. [Google Scholar] [CrossRef]
  31. Liu, X.; Xu, H. Does Low-Carbon Pilot City Policy Induce Low-Carbon Choices in Residents? Living: Holistic and Single Dual Perspective. J. Environ. Manag. 2022, 324, 116353. [Google Scholar] [CrossRef]
  32. Zheng, Y.; Zhang, M.; Wang, S.; Wang, L. The Impacts of Low-Carbon City Pilot Policies on Natural Population Growth: Empirical Evidence from China’s Prefecture-Level Cities. Front. Public Health 2023, 11, 1214070. [Google Scholar] [CrossRef]
  33. Wang, Z.; Cao, X.; Ren, X. Green Development and Economic Resilience: Evidence from Chinese Resource-Based Cities. Front. Eng. Manag. 2024, 11, 194–206. [Google Scholar] [CrossRef]
  34. Tan, J.; Zhang, P.; Lo, K.; Li, J.; Liu, S. Conceptualizing and Measuring Economic Resilience of Resource-Based Cities: Case Study of Northeast China. Chin. Geogr. Sci. 2017, 27, 471–481. [Google Scholar] [CrossRef]
  35. Wu, X.; Zhang, J.; Geng, X.; Wang, T.; Wang, K.; Liu, S. Increasing Green Infrastructure-Based Ecological Resilience in Urban Systems: A Perspective from Locating Ecological and Disturbance Sources in a Resource-Based City. Sust. Cities Soc. 2020, 61, 102354. [Google Scholar] [CrossRef]
  36. Chen, Y.; Wang, H. Industrial Structure, Environmental Pressure and Ecological Resilience of Resource-Based Cities-Based on Panel Data of 24 Prefecture-Level Cities in China. Front. Environ. Sci. 2022, 10, 885976. [Google Scholar] [CrossRef]
  37. He, M.; Xiao, W.; Zhao, L.; Xu, Y. Spatiotemporal Evolution Pattern and Heterogeneity of Resource-Based City Resilience in China. Struct. Chang. Econ. Dyn. 2024, 71, 417–429. [Google Scholar] [CrossRef]
  38. Tan, J.; Hu, X.; Hassink, R.; Ni, J. Industrial Structure or Agency: What Affects Regional Economic Resilience? Evidence from Resource-Based Cities in China. Cities 2020, 106, 102906. [Google Scholar] [CrossRef]
  39. Sun, Y.; Wang, Y.; Zhou, X.; Chen, W. Are Shrinking Populations Stifling Urban Resilience? Evidence from 111 Resource-Based Cities in China. Cities 2023, 141, 104458. [Google Scholar] [CrossRef]
  40. Ning, X.; Zhao, J.; An, Y. Urbanization and Urban Ecological Resilience in Resource-Based Cities: Coupling Coordination and Its Key Factors. Environ. Dev. Sustain. 2024. [Google Scholar] [CrossRef]
  41. Zhang, W.; Wang, Z.; Wang, S. The Impact of Land-Use Carbon Efficiency on Ecological Resilience-The Moderating Role of Heterogeneous Environmental Regulations. Sustainability 2024, 16, 9842. [Google Scholar] [CrossRef]
  42. Lu, M.; Tan, Z.; Yuan, C.; Dong, Y.; Dong, W. Resilience Measurements and Dynamics of Resource-Based Cities in Heilongjiang Province, China. Land 2023, 12, 302. [Google Scholar] [CrossRef]
  43. Feng, S.; Gao, B.; Tan, Y.; Xiao, K.; Zhai, Y. Resource-Based Transformation and Urban Resilience Promotion: Evidence from Firms’ Carbon Emissions Reductions in China. J. Clean. Prod. 2024, 468, 143118. [Google Scholar] [CrossRef]
  44. Wang, C.; Liu, X.; Li, H.; Yang, C. Analyzing the Impact of Low-Carbon City Pilot Policy on Enterprises’ Labor Demand: Evidence from China. Energy Econ. 2023, 124, 106676. [Google Scholar] [CrossRef]
  45. Li, S.; Zheng, X.; Liao, J.; Niu, J. Low-Carbon City Pilot Policy and Corporate Environmental Performance: Evidence from a Quasi-Natural Experiment. Int. Rev. Econ. Financ. 2024, 89, 1248–1266. [Google Scholar] [CrossRef]
  46. Ellis, H.; Zolotas, X. Economic-Growth and Declining Social-Welfare. J. Econ. Hist. 1985, 45, 770–772. [Google Scholar] [CrossRef]
  47. Zou, C.; Huang, Y.; Wu, S.; Hu, S. Does “Low-Carbon City” Accelerate Urban Innovation? Evidence from China. Sustain. Cities Soc. 2022, 83, 103954. [Google Scholar] [CrossRef]
  48. Zhang, Y.; Feng, N.; Wang, X. Can the Green Finance Pilot Policy Promote the Low-Carbon Transformation of the Economy? Int. Rev. Econ. Financ. 2024, 93, 1074–1086. [Google Scholar] [CrossRef]
  49. Ma, J.; Hu, Q.; Shen, W.; Wei, X. Does the Low-Carbon City Pilot Policy Promote Green Technology Innovation? Based on Green Patent Data of Chinese A-Share Listed Companies. Int. J. Environ. Res. Public Health 2021, 18, 3695. [Google Scholar] [CrossRef]
  50. Gu, R.; Li, C.; Li, D.; Yang, Y.; Gu, S. The Impact of Rationalization and Upgrading of Industrial Structure on Carbon Emissions in the Beijing-Tianjin-Hebei Urban Agglomeration. Int. J. Environ. Res. Public Health 2022, 19, 7997. [Google Scholar] [CrossRef]
  51. Yang, S.; Jahanger, A.; Hossain, M.R. How Effective Has the Low-Carbon City Pilot Policy Been as an Environmental Intervention in Curbing Pollution? Evidence from Chinese Industrial Enterprises. Energy Econ. 2023, 118, 106523. [Google Scholar] [CrossRef]
  52. Wang, D.; Chen, S. The Effect of Pilot Climate-Resilient City Policies on Urban Climate Resilience: Evidence from Quasi-Natural Experiments. Cities 2024, 153, 105316. [Google Scholar] [CrossRef]
  53. Shao, S.; Cheng, S.; Jia, R. Can Low Carbon Policies Achieve Collaborative Governance of Air Pollution? Evidence from China’s Carbon Emissions Trading Scheme Pilot Policy. Environ. Impact Assess. Rev. 2023, 103, 107286. [Google Scholar] [CrossRef]
  54. Cui, H.; Cao, Y. Energy Rights Trading Policy, Spatial Spillovers, and Energy Utilization Performance: Evidence from Chinese Cities. Energy Policy 2024, 192, 114234. [Google Scholar] [CrossRef]
  55. Wang, P.; Li, H.; Huang, Z. Low-Carbon Economic Resilience: The Inequality Embodied in Inter-Regional Trade. Cities 2024, 144, 104646. [Google Scholar] [CrossRef]
  56. Parkhill, K.A.; Shirani, F.; Butler, C.; Henwood, K.L.; Groves, C.; Pidgeon, N.F. “We Are a Community [but] That Takes a Certain Amount of Energy”: Exploring Shared Visions, Social Action, and Resilience in Place-Based Community-Led Energy Initiatives. Environ. Sci. Policy 2015, 53, 60–69. [Google Scholar] [CrossRef]
  57. Weitzel, E.C.; Glaesmer, H.; Hinz, A.; Zeynalova, S.; Henger, S.; Engel, C.; Loeffler, M.; Reyes, N.; Wirkner, K.; Witte, A.V.; et al. What Builds Resilience? Sociodemographic and Social Correlates in the Population-Based LIFE-Adult-Study. Int. J. Environ. Res. Public Health 2022, 19, 9601. [Google Scholar] [CrossRef]
  58. Zhang, Y.; Yang, Y.; Chen, Z.; Zhang, S. Multi-Criteria Assessment of the Resilience of Ecological Function Areas in China with a Focus on Ecological Restoration. Ecol. Indic. 2020, 119, 106862. [Google Scholar] [CrossRef]
  59. Han, H.; Gu, R.; Yang, Y. Impacts of Low-Carbon City Pilot Policy on Ecological Well-Being Performance across Chinese Cities: A Spatial Difference-in-Difference Analysis. Sustain. Cities Soc. 2025, 118, 105864. [Google Scholar] [CrossRef]
  60. Shi, C.; Guo, N.; Gao, X.; Wu, F. How Carbon Emission Reduction Is Going to Affect Urban Resilience. J. Clean. Prod. 2022, 372, 133737. [Google Scholar] [CrossRef]
  61. Li, G.; Liu, M. Spatiotemporal Evolution and Influencing Factors of Economic Resilience: Evidence from Resource-Based Cities in China. Sustainability 2022, 14, 10434. [Google Scholar] [CrossRef]
  62. Feng, Y.; Lee, C.; Peng, D. Does Regional Integration Improve Economic Resilience? Evidence from Urban Agglomerations in China. Sustain. Cities Soc. 2023, 88, 104273. [Google Scholar] [CrossRef]
  63. Xiong, B.; Sui, Q. Does Carbon Emissions Trading Policy Improve Inclusive Green Resilience in Cities? Evidence from China. Sustainability 2023, 15, 12989. [Google Scholar] [CrossRef]
  64. Xu, X.; Wang, M.; Wang, M.; Yang, Y.; Wang, Y. The Coupling Coordination Degree of Economic, Social and Ecological Resilience of Urban Agglomerations in China. Int. J. Environ. Res. Public Health 2023, 20, 413. [Google Scholar] [CrossRef]
  65. Shi, C.; Zhu, X.; Wu, H.; Li, Z. Assessment of Urban Ecological Resilience and Its Influencing Factors: A Case Study of the Beijing-Tianjin-Hebei Urban Agglomeration of China. Land 2022, 11, 921. [Google Scholar] [CrossRef]
  66. Pan, A.; Zhang, W.; Shi, X.; Dai, L. Climate Policy and Low-Carbon Innovation: Evidence from Low-Carbon City Pilots in China. Energy Econ. 2022, 112, 106129. [Google Scholar] [CrossRef]
Land 14 00147 g001
Figure 1. Study area pilot.
Figure 1. Study area pilot.
Land 14 00147 g001
Land 14 00147 g002
Figure 2. Kernel density of spatial differentiation of the socio-ecological resilience of resource-based cities, 2003–2022. (a) Socio-ecological resilience kernel density; (b) Economic resilience kernel density; (c) Social resilience kernel density; (d) Ecological resilience kernel density.
Figure 2. Kernel density of spatial differentiation of the socio-ecological resilience of resource-based cities, 2003–2022. (a) Socio-ecological resilience kernel density; (b) Economic resilience kernel density; (c) Social resilience kernel density; (d) Ecological resilience kernel density.
Land 14 00147 g002
Land 14 00147 g003
Figure 3. Parallel trend test results. The solid line represents the estimated dynamic policy effects over time, with the vertical dashed lines indicating the 99% confidence interval. The horizontal line (zero line) serves as the baseline to assess whether the policy effect differs significantly from zero, while the vertical dashed line on the x-axis marks the policy implementation time, establishing the baseline for the test.
Figure 3. Parallel trend test results. The solid line represents the estimated dynamic policy effects over time, with the vertical dashed lines indicating the 99% confidence interval. The horizontal line (zero line) serves as the baseline to assess whether the policy effect differs significantly from zero, while the vertical dashed line on the x-axis marks the policy implementation time, establishing the baseline for the test.
Land 14 00147 g003
Land 14 00147 g004
Figure 4. Placebo test results. The placebo test results are shown as follows: The solid line represents the kernel density estimation of the variable, reflecting the distribution of the estimated values. The dashed line marks the reference significance level (baseline at p-value = 0.10), while the blue dots represent the p-values for each estimated value, illustrating how significance levels change with the estimated values.
Figure 4. Placebo test results. The placebo test results are shown as follows: The solid line represents the kernel density estimation of the variable, reflecting the distribution of the estimated values. The dashed line marks the reference significance level (baseline at p-value = 0.10), while the blue dots represent the p-values for each estimated value, illustrating how significance levels change with the estimated values.
Land 14 00147 g004
Land 14 00147 g005
Figure 5. Variable differences before and after propensity score matching. This figure is used to assess whether the PSM effectively balances the distribution of covariates between the treatment and control groups. If the matched points are closer to zero, it indicates that the PSM has effectively improved the balance of covariates. The matched points in the figure are closer to zero, indicating a good matching effect.
Figure 5. Variable differences before and after propensity score matching. This figure is used to assess whether the PSM effectively balances the distribution of covariates between the treatment and control groups. If the matched points are closer to zero, it indicates that the PSM has effectively improved the balance of covariates. The matched points in the figure are closer to zero, indicating a good matching effect.
Land 14 00147 g005
Land 14 00147 g006
Figure 6. Local Moran’s Index. Using the plotting module in Stata, local Moran’s scatter plots of socio-ecological resilience for resource-based cities in 2003, 2009, 2015, and 2022 were drawn, to examine whether the results are concentrated along a straight line. (a) Localized Moran’s Index, 2003; (b) Localized Moran’s Index 2009; (c) Localized Moran’s Index 2015; (d) Localized Moran’s Index in 2022.
Figure 6. Local Moran’s Index. Using the plotting module in Stata, local Moran’s scatter plots of socio-ecological resilience for resource-based cities in 2003, 2009, 2015, and 2022 were drawn, to examine whether the results are concentrated along a straight line. (a) Localized Moran’s Index, 2003; (b) Localized Moran’s Index 2009; (c) Localized Moran’s Index 2015; (d) Localized Moran’s Index in 2022.
Land 14 00147 g006
Land 14 00147 g007
Figure 7. Distribution of socio-ecological resilience values for two different types of cities. (a) Socio-ecological resilience of HHM; (b) Socio-ecological resilience of MML.
Figure 7. Distribution of socio-ecological resilience values for two different types of cities. (a) Socio-ecological resilience of HHM; (b) Socio-ecological resilience of MML.
Land 14 00147 g007
Land 14 00147 g008
Figure 8. Policy implications for socio-ecological resilience in different types of pilot cities. Using double machine learning in Python, the impact coefficients of low-carbon policies are assessed. The plotting module is used to generate a chart of policy impact coefficients, with the vertical axis representing the impact coefficients, *, **, and *** correspond to significance levels of 10%, 5%, and 1%, respectively. (a) Impact of LCCP on the SESR of MML-cities; (b) Impact of LCCP on the SESR of HHM-cities.
Figure 8. Policy implications for socio-ecological resilience in different types of pilot cities. Using double machine learning in Python, the impact coefficients of low-carbon policies are assessed. The plotting module is used to generate a chart of policy impact coefficients, with the vertical axis representing the impact coefficients, *, **, and *** correspond to significance levels of 10%, 5%, and 1%, respectively. (a) Impact of LCCP on the SESR of MML-cities; (b) Impact of LCCP on the SESR of HHM-cities.
Land 14 00147 g008
Table 2. Descriptive statistics.
Table 2. Descriptive statistics.
VariableVariable NameSymbolObsMeanStd. Dev.MinMax
Explanatory variableSocio-ecological resilienceSESR22800.2620.0640.0960.498
Core explanatory variablesLow-carbon pilot policiesLCCP22800.1820.3861.0000.000
Control variableScience expendituresPes22800.0090.0100.0010.207
Average wageAwa228010.500.7017.07913.72
Green SupportGsu22800.0060.0040.0010.022
Population densityPud22807.9330.9003.2969.532
Road densityRap22802.5940.5380.6914.096
Urbanization rateUrb22803.8410.3662.1904.575
Mediator variableGreen financeGF22800.2770.1040.5250.504
Industrial transformation IS22800.2810.250−0.6872.699
Carbon emission efficiencyCE22800.3320.1320.0061.071
Table 3. Baseline regression analysis of the socio-ecological resilience of resource-based cities in low-carbon pilots.
Table 3. Baseline regression analysis of the socio-ecological resilience of resource-based cities in low-carbon pilots.
Benchmark RegressionSocial
Resilience		Economic
Resilience		Ecological
Resilience
Model (1)Model (2)Model (3)Model (4)Model (5)Model (6)Model (7)
LCCP0.042 ***
(12.68)		0.027 ***
(9.94)		0.006 ***
(4.72)		0.007 ***
(5.24)		0.004 **
(2.20)		0.006 **
(2.70)		0.026 ***
(6.38)
ControlsNoYESNOYESYESYESYES
CityNoNoYESYESYESYESYES
YearNoNoYESYESYESYESYES
_cons0.255 ***
(178.2)		−0.081 ***
(−7.43)		0.261 ***
(1076.9)		0.196 ***
(8.33)		0.168 ***
(5.64)		0.108 **
(2.57)		0.653 ***
(8.18)
N2280228022802280228022802280
Adj. R20.06550.42520.90550.90750.80980.93550.7912
Note: The standard error is shown in parentheses. Statistical significance is denoted by asterisks: **, and *** correspond to significance levels of 5% and 1%, respectively.
Table 4. Robustness check results: Models (8) to (10) exclude other policy interference for testing; Model (11) is the PSM-DID test; and Model (12) presents the DML test results.
Table 4. Robustness check results: Models (8) to (10) exclude other policy interference for testing; Model (11) is the PSM-DID test; and Model (12) presents the DML test results.
Exclusion of Other PoliciesPSM-DIDDML
Model (8)Model (9)Model (10)Model (11)Model (12)
LCCP0.009 *
(1.96)		0.008 *
(1.75)		0.008 *
(1.73)		0.009 *
(1.79)		0.009 **
(0.004)
KAPC0.005
(1.17)
GFRI 0.013
(1.31)
NEDP 0.002
(0.41)
Constant0.189 ***
(4.22)		0.194 ***
(4.29)		0.193 ***
(4.27)		0.154 ***
(2.80)		0.001 **
(0.001)
ControlsYESYESYESYESYES
CityYESYESYESYESYES
YearYESYESYESYESYES
Observations22802280228010892280
R-squared0.9140.9130.9130.9190.916
Note: The standard error is shown in parentheses. Statistical significance is denoted by asterisks: *, **, and *** correspond to significance levels of 10%, 5%, and 1%, respectively.
Table 5. Global Moran’s I. The global Moran’s index of socio-ecological resilience for resource-based cities in China from 2003 to 2022 was calculated using a spatial economic-geographic matrix.
Table 5. Global Moran’s I. The global Moran’s index of socio-ecological resilience for resource-based cities in China from 2003 to 2022 was calculated using a spatial economic-geographic matrix.
YearGlobal Moran’s Iz-ValueYearGlobal Moran’s Iz-Value
20030.174 ***3.28220130.303 ***5.677
20040.206 ***3.86220140.261 ***4.918
20050.131 **2.52020150.224 ***4.257
20060.210 ***3.95820160.185 ***3.532
20070.149 ***2.83720170.201 ***3.837
20080.252 ***4.69420180.222 ***4.196
20090.276 ***5.12920190.228 ***4.282
20100.235 ***4.41120200.230 ***4.326
20110.281 ***5.23520210.252 ***4.722
20120.285 ***5.31220220.262 ***4.905
Note: Asterisks indicate statistical significance, *** and ** represent 1% and 5% significance levels, respectively.
Table 6. SDID model test.
Table 6. SDID model test.
TestValueTestValue
LM-Spatial Lag29.774 ***LR-Spatial lag75.980 ***
Robust LM-Spatial Lag69.229 ***LR-Spatial Error74.3100 ***
LM-Spatial Error173.871 ***Wald-Spatial lag77.080 ***
Robust LM-Spatial Error213.325 ***Wald-Spatial Error75.4800 ***
Hausman558.090 ***
Note: Asterisks denote statistical significance, with *** indicating significance at the 1% levels. (Using Stata code to test the validity of the spatial model, the LM test is used to check for spatial effects, the Hausman test is used to determine whether the model uses fixed or random effects. The Wald test is used to assess the goodness of fit of the model, and the LR test is used to examine whether the Durbin model suffers from degeneration.).
Table 7. Spatial econometric estimation results of the impact of low-carbon pilots on socio-ecological resilience.
Table 7. Spatial econometric estimation results of the impact of low-carbon pilots on socio-ecological resilience.
SESRSocial ResilienceEconomic ResilienceEcological Resilience
Model (13)Model (14)Model (15)Model (16)
LCCP0.004 ***
(2.88)		0.002
(1.64)		0.003 **
(1.98)		0.015 ***
(2.60)
W × LCCP0.011 ***
(3.76)		0.012 ***
(3.78)		0.006 **
(2.05)		0.011
(0.94)
SR_Direct0.004 ***
(3.21)		0.003 *
(1.82)		0.003 **
(2.33)		0.015 ***
(2.81)
SR_Indirect0.016 ***
(4.86)		0.012 ***
(3.78)		0.007 **
(2.25)		0.012
(0.96)
SR_Total0.016 ***
(4.86)		0.015 ***
(4.33)		0.010 ***
(2.88)		0.027 **
(2.09)
LR_Direct0.012 ***
(2.83)		0.014
(1.50)		0.024
(0.12)		0.035 ***
(2.66)
LR_Indirect0.030 ***
(3.39)		0.070 ***
(2.93)		0.160
(0.18)		0.014
(0.60)
LR_Total0.042 ***
(4.59)		0.084 ***
(3.30)		0.184
(0.20)		0.049 **
(2.09)
ControlsYesYesYesYes
City FEYesYesYesYes
Year FEYesYesYesYes
Observations2160216021602160
R-squared0.9310.8870.9750.800
code114114114114
Note: The standard error is shown in parentheses. Statistical significance is denoted by asterisks: *, **, and *** correspond to significance levels of 10%, 5%, and 1%, respectively.
Table 8. Mechanism analysis: based on Equation (3).
Table 8. Mechanism analysis: based on Equation (3).
GFISCE
LR_Direct0.009 **
(2.20)		−0.030 **
(−2.07)		0.007
(0.82)
LR_Indirect0.063 ***
(3.42)		0.038
(1.41)		0.088 ***
(3.77)
ControlsYESYESYES
City FEYESYESYES
Year FEYESYESYES
Observations228022802280
R-squared0.1050.0010.003
Note: The standard error is shown in parentheses. Statistical significance is denoted by asterisks: ** and *** correspond to significance levels of 5% and 1%, respectively.
Table 9. Mechanism analysis: based on Equation (4).
Table 9. Mechanism analysis: based on Equation (4).
SESRSESRSESR
Direct
LCCP0.009 **
(2.20)		0.006 ***
(3.68)		0.006 ***
(3.57)
GF0.019 *
(1.93)
IS 0.005 **
(2.02)
CE 0.007 *
(1.82)
Indirect
LCCP0.063 ***
(3.42)		0.022 ***
(6.06)		0.023 ***
(6.33)
ControlsYESYESYES
City FEYESYESYES
Year FEYESYESYES
Observations228022802280
R-squared0.1050.0010.003
Note: The standard error is shown in parentheses. Statistical significance is denoted by asterisks: *, **, and *** correspond to significance levels of 10%, 5%, and 1%, respectively.
Table 10. Heterogeneity test: different types of resource-based cities.
Table 10. Heterogeneity test: different types of resource-based cities.
Growing CityMature CitiesDeclining CitiesRegenerative City
LCCP−0.002
(−0.17)		0.006 ***
(2.88)		0.018 ***
(6.21)		−0.007
(−1.29)
Constant0.121
(1.31)		0.257 ***
(9.20)		0.329 ***
(4.27)		−0.099
(−1.47)
ControlsYESYESYESYES
CityYESYESYESYES
YearYESYESYESYES
Observations2801240460300
R-squared0.9130.9230.8900.941
Note: The standard error is shown in parentheses. Statistical significance is denoted by asterisks: *** correspond to significance levels of 1%.
Table 11. Results and classification of the K-means clustering for SESR units.
Table 11. Results and classification of the K-means clustering for SESR units.
ClassificationEcological
Resilience		Economic
Resilience		Social
Resilience		City Name
MeanGradeMeanGradeMeanGrade
Category113.8H14.66H5.53MChizhou, Nanping, Huaibei, Guangyuan and 16 other cities
Category211.15M12.94M0.31LBaoji, Puer, Sanming, Shaoguan, Weinan and 20 other cities
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Peng, Y.; Wang, Z.; Zhang, Y.; Wang, W. Facilitating or Hindering? The Impact of Low-Carbon Pilot Policies on Socio-Ecological Resilience in Resource-Based Cities. Land 2025, 14, 147. https://doi.org/10.3390/land14010147

AMA Style

Peng Y, Wang Z, Zhang Y, Wang W. Facilitating or Hindering? The Impact of Low-Carbon Pilot Policies on Socio-Ecological Resilience in Resource-Based Cities. Land. 2025; 14(1):147. https://doi.org/10.3390/land14010147

Chicago/Turabian Style

Peng, Yanran, Zhong Wang, Yunhui Zhang, and Wei Wang. 2025. "Facilitating or Hindering? The Impact of Low-Carbon Pilot Policies on Socio-Ecological Resilience in Resource-Based Cities" Land 14, no. 1: 147. https://doi.org/10.3390/land14010147

APA Style

Peng, Y., Wang, Z., Zhang, Y., & Wang, W. (2025). Facilitating or Hindering? The Impact of Low-Carbon Pilot Policies on Socio-Ecological Resilience in Resource-Based Cities. Land, 14(1), 147. https://doi.org/10.3390/land14010147

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop