Analysis of the Spatial Effect of Carbon Emissions on Chinese Economic Resilience in the Context of Sustainability

Abstract

This paper focuses on the impact of carbon intensity on economic resilience in the context of the low-carbon development model and the promotion of sustainable development. Based on the provincial panel data from 2010 to 2021, this paper establishes a spatial econometric model to assess the impact of carbon intensity on economic resilience and applies the DID model to explore the impact of carbon emissions trading policies on economic resilience. It is found that carbon intensity suppresses the economic resilience of the local and associated regions. We also found through our research that carbon intensity can affect economic resilience through industrialization improvement and intensified pollution. In this situation, environmental regulatory policies are necessary to ensure sustainable development. The study found that the carbon emissions trading pilot policy could promote the input intensity in science and technology and technological manpower investments in the region, thus increasing the economic resilience. Moreover, the carbon emissions trading pilot policy is conducive to the economic resilience of neighboring regions. Based on the above research results, this paper proposes policy recommendations from three aspects: further promoting the upgrade of industrial structure and reducing carbon intensity; building a clean and low-carbon energy system to effectively reduce carbon and pollution emissions; and further improving the supporting measures for technological innovation to promote the economic resilience of developing countries and achieve sustainable development.

1. Introduction

Since the Industrial Revolution, the worldwide economy has evolved at an exponential rate. There has also been a significant improvement in people’s living standards and social modernization. At the same time, industrial development consumes fossil energy, resulting in the emission of carbon dioxide as a byproduct of this process. The total amount of carbon dioxide emissions, compared to that before the Industrial Revolution, has significantly increased, leading to a series of adverse effects on the ecological environment and global climate. Excessive carbon emissions can cause greenhouse effects and global warming, which are not conducive to environmental protection or sustainable development. From 2023 to 2027, global near-surface temperatures will be 1.1 to 1.8 °C higher than the pre-industrial revolution average, and there will be a further upward trend, posing many risks to terrestrial and marine ecosystems, human health, food and water security, economic and social de-elopements, etc. [1]. Therefore, carbon reduction and green development have gradually become the main way for countries around the world to address global climate issues. To control carbon dioxide emissions and delay the trend of global warming, countries around the world have successively created and signed the United Nations Framework Convention on Climate Change (1992), the Kyoto Protocol (1997), the Cancun Agreement (2010), and the Paris Agreement (2016). The 28th Conference of the Parties (COP28) to the United Nations Framework Convention on Climate Change (UNFCCC) in Dubai, UAE, came to a close on 13 December 2023 after two weeks of arduous negotiations. The conference completed its first global stocktaking, focusing on issues including emissions reduction and energy transition, adaptation and loss and damage, finance, technology, and capacity building support, and strengthening international cooperation. Given that the global climate change brought about by economic development has become a major challenge across countries around the world, the effective reduction in carbon emissions has become a key factor in the long-term and sustainable development of China and even the world’s economy and society. At the 75th United Nations Congress, the Chinese government solemnly announced that China aims to reach a carbon peak by 2030 and strive to achieve carbon neutrality by 2060. General Secretary Xi Jinping of the CPC Central Committee proposed in the report of the 20th CPC National Congress that China will actively and steadily promote carbon peaking and carbon neutrality, demonstrating China’s resolute determination to combat carbon emissions.

As the world’s largest developing country, China has always attached importance to environmental protection and economic security. China has long been promoting steady economic growth and focusing on strengthening ecological protection, actively minimizing environmental pollution while promoting economic growth and industrial development. In this regard, high-quality economic development is a better option. It refers to a stage of economic development with greater requirements, with a key issue of protecting the ecological environment. It is necessary to obtain a higher allocation efficiency, good economic development environment, and social benefits with lower resources and reduced environmental costs [2,3]. To achieve sustainable and high-quality economic developments, it is necessary to enhance the risk perception and impact resistance of the economic system, and ensure the sustained and healthy operation of the national economy by strengthening its economic resilience. At present, China’s economic development is confronted by significant issues, such as regional development disharmony and ecological environment [4]. Specifically, the top five provincial-level regions with the largest total carbon emissions in 2021 are Shandong, Hebei, Jiangsu, Inner Mongolia, and Guangdong. Ningxia, Inner Mongolia, Xinjiang, and Shanxi have the highest carbon intensity scores. According to the Global Air-Quality Report published by the World Health Organization, 47 cities in China were included in the top 100 air-polluting cities in the world in 2019, and China’s atmospheric environmental quality ranked 96 out of 117 countries on the list in 2021. Air pollution prevention and control have become common goals of China’s ecological civilization and important parts of high-quality developments in the process of Chinese-style modernization. To that end, the Action Plan for Carbon Peaking before 2030 states that all regions should take into account the actual economic and social developments and resource and environmental endowments of the region, and adhere to classified measures and promote peak carbon goals in a gradual and orderly manner.

Carbon emissions trading is one of the important tools to achieve the goals of carbon peaking and carbon neutrality, and the implementation of carbon emissions reduction policies can help enhance economic resilience. Since the completion of the European Union carbon emissions trading market in 2005, the carbon emissions trading mechanism has been implemented in 29 countries and regions around the world. China has also established a carbon emissions trading system at the national or local levels and has made important progress. In October 2011, the National Development and Reform Commission (NDRC) identified six provinces or municipalities, namely, Guangdong, Hubei, Beijing, Tianjin, and Chongqing, as well as Shenzhen, as the seven pilot regions for carbon emissions trading, of which Shenzhen was the first to open the market in June 2013, and the rest of the provinces opened online trading in succession in 2013 and 2014. On 16 July 2021, the national carbon market officially launched online trading, becoming the largest carbon emissions trading system in the world covering carbon emissions.

There are close development connections between regions, and changes in carbon emissions can have spatial effects on the economic resilience of the associated regions. In this regard, this paper discusses the impacts of carbon intensity and carbon emissions policies on economic resilience, as well as its internal factors.

To promote sustainable development, we construct a spatial econometric model to assess the impact of carbon intensity on economic resilience based on the provincial panel data in 2010–2021. Meanwhile, we apply the DID model to explore the impact of carbon emissions trading policies on economic resilience. It is found that carbon intensity suppresses the economic resilience of the local and associated regions. We also find that carbon intensity can affect economic resilience through industrialization improvements and intensified pollution. In this case, environmental regulatory policies are necessary to guarantee sustainable development. The study found that the carbon emissions trading pilot policy could promote the input intensity in science and technology and technological manpower investment intensity in the region, thus increasing economic resilience. Moreover, the carbon emissions trading pilot policy is conducive to the economic resilience of neighboring regions. Based on the above findings, we put forward relevant policy recommendations to facilitate developing countries to improve their economic resilience and achieve sustainable development.

For the remaining parts of this paper, Section 2 presents a literature review; Section 3 is a theoretical analysis of the impact mechanism; Section 4 is an introduction to the Materials and Methods; Section 5 reports the empirical conclusions and related analysis; Section 6 is a further analysis of the policy impacts; and the Section 7 is a discussion of the conclusions and implications.

2. Literature Review

2.1. Research on Economic Resilience

Fujita and Thisse [5] first introduced resilience into the field of spatial economics re-search, referring to the multiple equilibrium approach in the ecological resilience theory to explain the internal mechanisms of various agglomeration phenomena in the real economy. Then, the research on economic resilience gradually began. Scholars’ explorations of economic resilience cover two perspectives: short-term economic resilience and long-term economic resilience. Due to the different perspectives of the analyses, scholars also presented differences in measuring economic resilience and analyzing the influencing factors. Martin [6] views economic resilience as the capacity of a region to resist and recover from crises and argues that shocks will not have a long-term impact on the development of a region. In this regard, scholars suggest that crises lead to economic losses. When measuring economic resilience, one or more indicators are usually selected to calculate the economic losses and sensitivity before and after a crisis to reflect the economic resilience outcome [6,7,8]. Some scholars have also constructed indicator systems from the perspectives of resistance, resilience, and adaptability to measure economic resilience [9]. Di Pietro et al. [10] analyzed the economic system’s adjustment and recovery mechanisms under various impact scenarios using a spatial general equilibrium model. Glaeser [11] and Crespo et al. [12] proposed that factors, such as total factor productivity, market size, openness, marketization, human capital, and innovation ability, would affect the resilience of regional economies to resist fluctuations and recover from crises. Simmie et al. [13], Martin [14], and Boschma [15] argued that regional economies are constantly undergoing minor shocks and are in a dynamic development process. Therefore, economic resilience is the long-term ability of regional economies to continuously adjust their economic structure and scale to adapt to external shocks and achieve economic development. Brigglio et al. [9] constructed comprehensive indicators of economic resilience from four aspects: market efficiency, governance efficiency, social development, and economic stability. In accordance with Davies [16], Brown et al. [17], Fingleton et al. [18], and Huggins [19], industrial structure, economic development level, and cultural features constitute significant determinants that impact economic resilience.

2.2. Research on Carbon Emissions

Scholars from various countries have extensively discussed carbon emissions. On the one hand, relevant scholars have conducted a lot of research on the spatiotemporal characteristics [20] and driving factors [21] of carbon emissions. Many scholars’ studies focus on the impact of factors, such as trade openness, industrial structure upgrading, economic structure adjustment, foreign investment introduction, and regional integration, on the total amount and intensity of carbon emissions [22,23,24,25,26]. Ang et al. [27] discovered that higher industrial productivity resulted in increased carbon emissions using the LMDI index decomposition approach. Liu L [28] found that, from the perspective of total factor carbon emissions efficiency, China’s urbanization rate had a positive promoting effect on the improvement of the total factor carbon emissions efficiency, and there was a positive spatial spillover effect. According to Acheampong et al. [29], the influence of financial development on carbon intensity varies at different stages of development, with sophisticated financial economies having the greatest carbon reduction effect. Timilsina and Shrestha [30] used an exponential decomposition method to investigate the factors influencing the carbon intensity of the transportation sector in 20 Latin American and Caribbean countries from 1980 to 2005. The findings revealed that economic growth, carbon coefficients, and energy intensity had the greatest impacts on carbon intensity. Economic development has been accompanied by an increase in carbon emissions, and the ensuing environmental problems have raised concerns. Liao et al. [31] found that environmental pollution exacerbated income-related health inequalities, particularly affecting middle-aged people. Carbon emissions will also increase the probability of extreme weather events, such as droughts, floods, and hail [32].

On the other hand, scholars have analyzed the impacts of environmental policies and regulations on carbon emissions. At the provincial level, Wang et al. [33] found through testing that the pilot policy for carbon emissions permit trading had achieved certain results, significantly reducing the carbon emissions of the pilot areas. At the urban level, Yu and Zhang [34] and Liu et al. [35] examined the carbon and pollutant emissions effects of low-carbon city pilot policies and innovative city pilot policies on cities. Porter and Linde [3] investigated the influence of environmental regulations on urban carbon intensity and concluded that both environmental policy and environmental regulation had a dampening effect on urban carbon emissions. Lin and Zhu [36] and Zhang et al. [37] used DID and PSM-DID methods to specifically evaluate the impact of policies, such as low-carbon city pilot projects and carbon emissions permit trading, and analyzed the spatiotemporal effects of carbon emissions policies on carbon emissions reductions. Yu et al. [34] found through a data envelopment analysis that low-carbon pilot policies had significantly improved the carbon emissions efficiency of pilot cities, and there was a positive spatial spillover effect. Chen and Lin [38] used the global DEA and SCM models to validate the emissions reduction effectiveness of carbon emissions trading policies, demonstrating that China’s carbon emissions trading market, while still in its early stages of development, had a significant impact on promoting carbon emissions reduction and improving carbon performance.

In addition, scholars have conducted research on the calculation and decomposition of carbon intensity. Wang et al. [39], Xing et al. [40], Uvarova and Kuzovkin [41], and GEA [42] measured the total carbon emissions, industrial carbon emissions, and service industry carbon emissions based on diversified carbon emissions estimation methods. They found that cities consume 60–80% of global energy and emit 70% of global greenhouse gases. Wang Z et al. [43] used methods, such as the super-efficiency SBM model, kernel density estimation, and geographic detectors, to measure and analyze the spatiotemporal differences in transportation carbon emissions efficiency between the Yangtze River Economic Belt and the Yellow River Basin. Clarke-Sather et al. [44] classified China into three distinct areas, eastern, central, and western, and subsequent to that, analyzed the coefficient of variation, Gini coefficient, and Theil index to analyze the variations between regions in carbon intensity. By analyzing the abovementioned literature, this study constructs a spatial econometric model to assess the impact of carbon intensity on economic resilience and the mechanisms behind it. This study further applies the DID model to explore the impact of carbon emissions trading policies on economic resilience.

The possible marginal contributions of this study are mainly reflected in two aspects. Firstly, it is more reasonable than traditional methods for evaluating influencing factors and policy effectiveness. This study not only evaluates the impacts of carbon intensity and carbon emissions trading pilot policies on the economic resilience of the local area, but also evaluates the impacts of carbon intensity and carbon emissions trading pilot policies on the economic resilience of neighboring areas by creating a spatial econometric model. Secondly, this study broadens the perspective of the existing literature. In the context of sustainable development, this study considers industrial structure, environmental pollution, and technological innovation, and verifies the theoretical mechanism of the impact of carbon intensity on economic resilience through the “degree of industrialization” and “degree of pollution”. Meanwhile, this study also verifies the theoretical mechanism of the impact of carbon emissions trading pilot policies on economic resilience through two aspects: “input intensity in science and technology” and “technology manpower investment”. WE also provide microlevel evidence to demonstrate that carbon intensity suppresses the economic resilience of the local and associated regions, and that carbon emissions trading pilot policies can promote the Improvement of economic resilience in the local and associated regions.

3. Influencing Mechanism

The acceleration of regional urbanization has promoted the high level of industrialization. Energy-consuming industries in regions rely on a large amount of energy and resource consumption to achieve rapid development, thereby releasing the dividends of economies of scale, attracting a large number of people to gather, promoting the expansion of regional industrial sectors, and increasing regional carbon emissions and industrial carbon intensity. The increasing intensity of carbon emissions poses challenges to the green development of a local economy. From the point of view of green economy development, decreasing the carbon intensity can promote high-quality local economic development, stabilize the economic growth environment, and strengthen economic resilience. Secondly, the development of the regional economy also benefits from the agglomeration effect created by industrial agglomeration. The new production cooperation mode of the cross-industry division of labor and collaboration, resource and energy information sharing among different enterprises, has improved the production efficiency of enterprises in the region. At the same time, the development of this region also plays a fundamental and leading role in economically related regions, prompting them to follow suit. The local and associated regions will subsequently modify the growth of their economic models, adjust their industrial structures, encourage the development of the region’s green economy, and lead to the regional economy’s long-term development. Therefore, from the perspective of economic development, a decrease in carbon intensity will have a spatial spillover effect on economically related regions.

Hypothesis 1.

The decreased carbon intensity is convenient when it comes to enhancing local economic resilience as well as strengthening economic resilience in economically related regions with spatial spillover effects.

An industrial structure with high energy consumption and severe environmental pollution levels, such as high carbonization energy consumption structures dominated by the use of coal, and high energy consumption intensity are important factors in the generation of carbon dioxide. High carbon intensity is frequently associated with a high level of industrialization and pursuing the improvement of the industrialization level impedes the upgrading of industrial structures. The allocation of resources in a region relies on the industrial structure, and the process of upgrading the industrial structure is also the process of appreciating resources. The upgrading of industrial structure can break the market segmentation situation, promote the effective utilization of various production resources between industries and internal economic departments, achieve industrial chain interconnections, enhance economic stability, and enhance regional economic resilience. The upgrading of the industrial structure means reducing the proportions of primary and secondary industries, and the improvement of the industrialization level inevitably affects the upgrading of the industrial structure, thus affecting economic resilience. Therefore, carbon intensity is closely related to the level of industrialization, and the improvement of the industrialization level is not conducive to the improvement of regional economic resilience.

Hypothesis 2.

The level of industrialization plays a mediating role between carbon intensity and economic resilience.

Ecological environmental pollution, global climate change, and other issues are closely related to the increase in greenhouse gas emissions. Carbon intensity is an essential factor that affects the quality of the ecological environment, which can have an impact on economic resilience. Natural resources are the material guarantee for development, and ecological pollution limits the natural resources needed for production and life, thereby reducing the effective utilization rate of resources. Meanwhile, the deterioration of the ecological environment has led to an increase in the cost of public healthcare and an increase in the risk of the net loss of social labor. Due to the higher education level, people often require a better quality of life, which directly affects the employment of high-quality talent and retention of existing high-tech talent in the region, and indirectly affects the accumulation of human capital and the overall technological level in the region. In addition, the quality of the ecological environment is closely related to social life. Climate changes, such as air pollution, can seriously affect the quality of life of people, greatly exacerbate regional pollution, and affect the working hours and efficiency of workers from both physiological and psychological aspects, thereby affecting the overall social efficiency and leading to a decrease in economic resilience.

Hypothesis 3.

The degree of pollution plays a mediating role between carbon intensity and economic resilience.

After years of growth, the Chinese economy has ranked second in the world in terms of economic scale. However, due to the constraints of the economic development mode and stage, problems, such as intensified environmental pollution and the deterioration of the ecological environment, have emerged. Currently, environmental pollution, represented by carbon emissions, has become a bottleneck for high-quality economic development in China and even the world. Low carbon emissions and green developments are imperative. Therefore, in order to reduce the environmental pollution, promote regional carbon emissions reductions, promote low-carbon economic developments, and achieve high-quality economic developments, the country has formulated a series of environmental protection policies, such as low carbon emissions, to accelerate the transformation of the economic development mode in a region, and thereby enhance the regional economic resilience.

After the implementation of carbon emissions policies, the carbon dioxide emissions of enterprises in the region will be limited by the emissions. To ensure their production scale, enterprises need to reduce carbon dioxide emissions through technology. In order to obtain greater profits, enterprises will carry out production equipment transformations, update their environmental protection equipment, increase technology investments, and recruit technology talents for technological innovations. On the one hand, investments in scientific and technological manpower in enterprises can enhance their human capital, improve the efficiency of production equipment use, improve productivity, and reduce unit output pollutant emissions. On the other hand, scientific and technological manpower investments can expand knowledge and technology spillovers through a talent exchange between regions, which is conducive to technological innovations for enterprises. With the transformation of production equipment and technological innovation, the production efficiency of enterprises has improved, leading to an increase in the energy utilization efficiency. Therefore, with the promotion of policies, innovation activities of enterprises in the region have increased; pollution emissions in the region have decreased; emissions of carbon dioxide, sulfur dioxide, industrial wastewater, etc., have decreased; and the contribution of technological innovation to the economy has increased, resulting in environmental and productivity improvements, promoting the resilience of the regional economy.

Hypothesis 4.

Carbon emissions policies can promote economic resilience by increasing the intensity of technology investment.

Hypothesis 5.

Carbon emissions policies can promote economic resilience by increasing the intensity of the use of technology personnel.

4. Research Design

4.1. Model Building

Spatial econometric models can be categorized into three types: spatial autoregressive models (SARs), spatial error models (SEMs), and spatial Durbin models (SDMs). The spatial Durbin model considers whether the spatial effects of an explained variable depend on the explanatory variables of the region and neighboring regions, which means that it has a more general form than the spatial autoregressive model and the spatial error model, and is consequently more effective at establishing the spillover effects between regions. In this paper, the SDM model was applied to explore the spatial effect of carbon intensity on economic resilience. Therefore, the regression model is as follows:

$$Re{s}_{it}=\rho WRe{s}_{it}+{\alpha }_{1}C{I}_{it}+{\alpha }_{2}WC{I}_{it}+{\alpha }_3{X}_{it}+{\mu }_i+{\gamma }_t+{\epsilon }_{it}$$

(1)

where i and t represent the region and time, respectively; $Re{s}_{it}$ is the predicted variable representing the economic resilience of region i at time t; $C{I}_{it}$ is the explanatory variable representing the carbon intensity of region i at time t; W is the spatial weight matrix; $X_{it}$ is the controlled variable; $\rho$ is the spatial regression coefficient (reflecting the spatial spillover effect); ${\mu }_i$ is an individual fixed effect; ${\gamma }_t$ is a fixed time effect; and ${\epsilon }_{it}$ is a random error term. ${\alpha }_2$ is the coefficient of the spatial lagged variable, which demonstrates the direction of action of the explanatory variable of the area surrounding it on the local predicted variable: when ${\alpha }_2$ > 0, the explanatory variables of adjacent regions have a significant positive spillover effect on the local predicted variables; when ${\alpha }_2$ < 0, the explanatory variables of adjacent regions have a significant negative spillover effect on the local predicted variables.

4.2. Variable Declaration

4.2.1. Economic Resilience

In terms of the economic resilience, referring to Chen [45] and the economic resilience measurement method proposed by Martin [6] and Martin and Moodysson [46], the change in employment in each region is used to measure this. The calculation formula is:

$$Re{s}_{it}=\frac{\left(L_{i,t}-L_{i,t-1}\right)/L_{i,t-1}-\left(L_{r,t}-L_{r,t-1}\right)/L_{r,t-1}}{\left|\left(L_{r,t}-L_{r,t-1}\right)/L_{r,t-1}\right|}$$

(2)

where $L_{i,t}$ and $L_{i,t-1}$ are the number of employees in each province during the t and t − 1 periods, while $L_{r,t}$ and $L_{r,t-1}$ are the number of employees in the country during the t and t − 1 periods. Res_it > 0 indicates that the larger the numerical value, the stronger the economic resilience; when Res_it < 0, the larger the absolute value, the stronger the economic resilience.

4.2.2. Carbon Intensity

Carbon dioxide (CO₂) emissions are primarily caused by the use of fossil fuels and combustion. The measurement of CO₂ emissions generally adopts the coefficient method or material balance method. The carbon emissions coefficient utilized was applied in this study to calculate the total CO₂ emissions of energy consumption in various regions, illustrating the inspiration from the methods of IPCC [47]: $Q_{C{O}_2}={\displaystyle \sum _{i=1}^n{E}_i\times S{C}_i}\times C{F}_i$ where Q_CO2 represents the total CO₂ emissions, $E_i$ represents the energy consumption of energy source i, $S{C}_i$ represents the conversion coefficient of standard coal, and $C{F}_i$ represents the carbon emissions coefficient of the energy source i. The specific carbon source composition and corresponding carbon emissions coefficients are shown in

Table 1. Carbon emissions coefficient and standard coal conversion coefficient.

Energy Type	Coal	Coke	Crude Oil	Gasoline	Kerosene	Diesel Oil	Fuel Oil	Natural Gas
Carbon emissions coefficient	0.7143	0.9714	1.4286	1.4714	1.4714	1.4571	1.4286	1.3300
Conversion coefficient of standard coal	1.9003	2.8604	3.0202	2.9251	3.0179	3.0959	3.1705	2.1622

Note: the measurement unit for the carbon emissions coefficient is kilogram standard coal/kilogram, the measurement unit for natural gas in the standard coal conversion coefficient is kilogram standard coal/cubic meter, and the measurement unit for other energy is kilogram standard coal/kilogram. The data are obtained from the General rules for calculation of the comprehensive energy consumption (GB/T 2589-2008) [48], and the Guidelines for the Preparation of Provincial Greenhouse Gas Inventories (FGBQH (2011) No. 1041) and are calculated.

. Calculate the carbon intensity based on the ratio of the total carbon emissions in each province to GDP in the same year after calculating the total carbon emissions in each region.

4.2.3. Mechanism Variables

(1)

Degree of industrialization. The degree of industrialization was measured by the proportion of secondary industry, which was expressed as the proportion of secondary industry added value to the gross domestic product in each province and city, and the logarithm of the proportion was also calculated.

(2)

Pollution level. The degree of pollution was represented by the complete investment amount in industrial pollution control in each province and city, and the logarithmic value of the proportion was also calculated.

(3)

The input intensity in science and technology. This was measured by the intensity of the R&D investment, expressed as the ratio of R&D investment in various provinces and cities to GDP, and the logarithm of the proportion was also calculated.

(4)

Technological manpower investment. The full-time equivalents of R&D staff in various provinces and cities were used to indicate the investments in scientific and technological manpower, and the logarithmic value of the proportion was also calculated.

4.2.4. Controlled Variables

(1)

Infrastructure. The infrastructure was represented by the proportion of the total postal and telecommunications businesses of each province and city to the gross domestic product, and the logarithm of this proportion was calculated.

(2)

Fixed-assets investment. Fixed-assets investments were expressed by the proportion of the total social fixed-assets investments in the GDPs of each province and city, and the logarithm of this proportion was calculated.

(3)

Urban population density. The urban population density was expressed as the ratio of the urban population in each province and city to the land area of the administrative region, and the logarithm of this proportion was calculated.

(4)

Degree of foreign trade. The degree of foreign trade was expressed as the proportion of the import and export amounts of goods in each province and city to the gross domestic product, and the logarithm of this proportion was calculated.

4.3. Data Sources

This study examined 30 provinces and cities in Mainland China (excluding Tibet) from 2010 to 2021 as the research samples. The raw data for each variable were obtained from the National Bureau of Statistics’ official website, China Statistical Yearbook, China Energy Statistical Yearbook, and the province and city statistical yearbooks.

Table 2. Descriptive statistics.

Variable Type	Variable	Number of Samples	Average	Standard Deviation	Min.	Max.
Explained variable	Res (economic resilience)	360	2.435	31.832	−187.388	367.569
Explanatory variables	CI (carbon intensity)	360	2.204	2.902	0.221	28.385
Mechanism variables	Ind (degree of industrialization)	360	3.745	0.424	2.388	6.713
	Pol (pollution level)	360	2.374	1.251	−3.045	4.953
	Itec (input intensity in science and technology)	360	0.015	0.103	−0.294	0.757
	Manp (technological manpower investment)	360	8.891	4.178	7.054	13.472
Controlled variables	Inf (infrastructure)	360	−2.752	1.106	−4.244	0.835
	Fix (fixed-assets investment)	360	0.354	1.154	−1.442	4.786
	Urb (urban population density)	360	7.854	0.714	6.639	8.669
	Fore (degree of foreign trade)	360	−3.653	1.061	−7.913	−1.02

Note: own compilation.

displays the descriptive statistics for each variable.

5. Empirical Analysis

5.1. Spatial Correlation Test

Spatial correlation refers to the mutual influence of various variables in space. The Moran index test is the most common method used to detect spatial correlations at present. This study selected the nearest neighbor weight matrix to calculate the global Moran index of carbon intensity and economic resilience from 2010 to 2021, as shown in

Table 3. Calculation results of the global Moran index.

Year	Moran Index
	Economic Resilience	CI (Carbon Intensity)
2010	0.176 (2.736) *	0.339 (2.521) **
2011	0.287 (2.172) **	0.354 (2.664) ***
2012	0.364 (2.719) ***	0.343 (2.625) ***
2013	0.439 (3.282) ***	0.348 (2.683) ***
2014	0.468 (3.476) ***	0.326 (2.527) **
2015	0.468 (3.356) ***	0.297 (2.325) **
2016	0.250 (2.834) ***	0.295 (2.270) **
2017	0.210 (2.321) *	0.302 (2.344) **
2018	0.689 (5.879) ***	0.221 (2.870) *
2019	0.422 (3.114) ***	0.227 (2.850) *
2020	0.154 (2.142) *	0.218 (2.800) *
2021	0.131 (2.265) *	0.216 (2.744) *

Note: ***, **, and * represent significance levels of 1%, 5%, and 10%, respectively. Own compilation.

. The significance levels of various indicators in Table 3 were tested, indicating a significant spatial correlation between carbon intensity and economic resilience, which could be estimated in the regression model for spatial effects.

This study categorized economic resilience and carbon intensity into four groups utilizing the natural breakpoint categorization method, as shown in Figure 1 and Figure 2. The categories that followed were low level, medium level, high level, and higher level. Figure 1 exhibits the distribution of carbon intensity in Chinese provinces with regard to 2010 and 2021. The intensity of carbon emissions was distributed in a pattern of “high in the north and low in the south” throughout the sample period. Inner Mongolia, Shaanxi, Liaoning, and Xinjiang have always been areas with high carbon intensities, while the central and southern provinces of Hunan, Jiangxi, Guangxi, Guangdong, Fujian, Zhejiang, and Jiangsu are areas with low carbon intensities. Specifically, from 2010 to 2021, the carbon intensities in Inner Mongolia and Shaanxi transitioned to high-level areas, while Gansu, Qinghai, Henan, Hubei, and Anhui transitioned to low-level areas. The vast majority of Chinese provinces’ carbon intensity levels demonstrated an increased likelihood of transitioning from high to low levels during the sample period.

Figure 2 depicts the distribution of economic resilience in China’s provinces between 2010 and 2021. Northeast China, Xinjiang, and Guangdong had comparatively strong economic resilience results in 2010, whereas other provinces had a relatively low overall level. Since then, China’s provinces’ and regions’ economic resilience levels have generally improved as the government pushed forward with accelerating the transformation of the mode of economic development, placing an emphasis on the construction of a resource-saving and environmentally friendly society as an important focus point, and promoting China’s economy’s high-quality development. At the present stage, Guangdong, Shanghai, Zhejiang, Hubei, and other regions are in high-value areas, while the economic resilience of the northeast region shows a downward trend.

From Figure 1 and Figure 2, it can be seen that, in the early stage, relying on energy consumption to promote economic development, regions with higher carbon intensity levels also have greater economic resilience. There are some regional variations in the distributions of carbon emissions intensity and economic resilience over time. That is, regions with greater economic resilience have lower carbon intensity levels, while regions with less economic resilience have higher carbon intensity levels. Both gradually exhibit a reverse correlation feature of “one decreasing and the other increasing” in space.

5.2. Empirical Study on the Spatial Effect of Carbon Intensity on Economic Resilience

This study used the SDM model (Formula (1)) based on the provincial panel data from 2010 to 2021 to analyze the impact of carbon intensity on economic resilience. Based on the results of the LM, LR, and Hausman tests, as well as the principle of the optimal fit, this study selected a bidirectional fixed-effect spatial Durbin model for the empirical analysis to estimate the spatial effect of carbon intensity on economic resilience. The specific regression results are shown in

Table 4. Spatial-effect estimation results.

Variable Type	Model (1)	Model (2)	Model (3)	Model (4)
				Direct Effect	Indirect Effect	Total Effect
CI (carbon intensity)	−2.686 **	−1.607 **	−4.136 ***	−4.136 ***	−33.430 *	−37.565 *
	(1.168)	(0.797)	(1.302)	(1.350)	(18.517)	(19.170)
W × CI	6.108	−16.507 *	−33.591 **
	(5.334)	(8.965)	(17.027)
Controlled variable	Yes	Yes	Yes	Yes
Fixed area	Yes	No	Yes	Yes
Fixed time	No	Yes	Yes	Yes
ρ	0.405 ***	−0.166	0.419 ***
	(0.121)	(0.213)	(0.198)
Number of samples	360	360	360	360	360	360
R2	0.001	0.002	0.006	0.006	0.006	0.006
Log-L	−103.6217	−123.9528	−194.6941	−194.6941

Note: (1) ***, **, and * represent significance levels of 1%, 5%, and 10%, respectively; (2) the values in parentheses in models (1)–(4) are standard errors. Own compilation.

.

Models (1)–(3) in Table 4 represent the estimated results of the regional fixed effects, time fixed effects, and bidirectional fixed effects, respectively. Model (4) is the estimated result of decomposing the effects of model (3) according to its influence through the partial differentiation method. The direct effect represents the impact of local carbon intensity on local economic resilience; the indirect effect represents the impact of local carbon intensity on the economic resilience of adjacent regions.

In accordance with the estimation results of model (4), the decomposition results of the effects of carbon intensity pass the significance test, and the estimated coefficients of the direct, indirect, and total effects are all negative, indicating that the greater the intensity of carbon emissions in the region, the less favorable the results are to the enhancement of economic resilience. The negative spatial spillover effect is quite significant, not only inside the region, but also where it shows an exponential increase in the geospatial distance, and the intensity of carbon emissions can pose a threat to the economic resilience across the regions. On the one hand, reducing the carbon intensity will promote the resilience of the local economy. The development of a green economy can better promote the high-quality development of the local economy, and regional economic development often relies mainly on high-energy-consuming industries. In order to ensure regional sustainable development, industrial structure adjustments are necessary. Technological innovations in the process of industrial structure upgrades helps to efficiently utilize production factors, expand the production scale, increase the labor demand, and facilitate regional economic constructions [49]. In addition, emerging industries born from industrial structure adjustments will create new job positions and employment needs, expand the employment stock and increment scale, attract more high-quality talents to enter the local area, strengthen regional development momentum, and promote economic resilience [50]. On the other hand, the reduction in carbon intensity is conducive to the protection of the regional ecological environment, the sustainable development of the local economy, and the balance of regional economic development. The radiative power of regional development can drive the construction of related regions. Affected by the industrial agglomeration effect, the cross-regional division of labor, cooperation, and information sharing among various industries and enterprises drive the adjustment and transformation of industrial development methods in the surrounding areas [51]. In addition, pollution control technologies can also have a scaled benefit-increasing effect. To reduce governance costs, surrounding areas will rely on areas with low-carbon-emissions reduction technologies to enhance the industrial connectivity between regions, gradually reducing carbon intensity levels in related areas, achieving a green transformation in economic development areas, and improving economic resilience [52,53]. This result validates Hypothesis 1 in this study.

5.3. Robust Test

This study tested the robustness of the estimation results from three aspects: replacing variables, replacing spatial weight matrices, and simultaneously replacing variables and spatial weight matrices. The estimation results are shown in

Table 5. Robustness test results.

Variable Type	Model (1)	Model (2)	Model (3)
CI (carbon intensity)	−5.346 ***	−3.239 **	−4.391 ***
	(1.680)	(1.269)	(1.659)
W × CI	−47.021 **	−131.618 ***	−178.167 ***
	(22.767)	(11.154)	(82.136)
Controlled variable	Yes	Yes	Yes
Spatial weight matrix	Adjacent matrix	Economic weight matrix	Economic weight matrix
ρ	0.411 **	−0.656 ***	−0.662 ***
	(0.199)	(0.172)	(0.181)
Number of samples	360	360	360
R2	0.006	0.007	0.007
Log-L	−194.5497	−199.2991	−199.0664

Note: (1) *** and ** represent significance levels of 1% and 5%, respectively; (2) the values in parentheses in models (1)–(3) are standard errors. Own compilation.

.

The estimation results of all the models in Table 5 are consistent with the original model, which means the robustness results of the abovementioned three aspects all demonstrate the fact that the estimation methods employed in this investigation are robust and reliable. In terms of the variables that were replaced, this paper substitutes the main explanatory variable with the carbon intensity deduced using coal energy, and the estimated the results are shown in model (1) in Table 5. In terms of transforming the spatial weight matrix, this study showed that the impact of carbon intensity on economic resilience was not only related to the distance between regions, but also to the regional economy. Therefore, this study calculated estimates based on the economic weight matrix, and the estimation results are shown in model (2). In terms of simultaneously replacing the variables and spatial weight matrix, this study replaced the core explanatory variables and spatial weight matrix according to models (1) and (2), and the estimated results are shown in model (3).

5.4. Mediating-Effect Test

The findings above indicate the fact that the intensity regarding carbon emissions has a detrimental influence on economic resilience enhancement. To investigate the underlying causes of this phenomenon, this study used the degree of industrialization and pollution as mediating variables to investigate the mechanism of carbon intensity influencing economic resilience, and it employed bootstrap statistical values for secondary testing to confirm the robustness of the mediating-effect test results. This study used the same regression method as in Formula (1), with the predicted variables in models (2) and (4) representing the industrialization and pollution levels, respectively. Models (3) and (5) added the industrialization and pollution levels as explanatory variables on the basis of Formula (1). The core explanatory variable for carbon intensity and controlled variables remain unchanged.

Table 6. Mediating-effect test results.

Variable Type	Model (1)	Model (2)	Model (3)	Model (4)	Model (5)
		Ind (Degree of Industrialization)		Pol (Pollution Level)
CI (carbon intensity)	−4.136 ***	0.074 ***	−2.252	0.047 *	−4.013 ***
	(1.302)	(0.005)	(1.654)	(0.025)	(1.277)
Ind (degree of industrialization)			−25.166 *
			(13.906)
Pol (pollution level)					−19.418 **
					(9.104)
Bootstrap test (indirect effect)		−1.654 ***		−0.256 *
		(0.608)		(0.150)
Bootstrap test (direct effect)		1.623 **		0.225
		(0.706)		(0.526)
Controlled variable	Yes	Yes	Yes	Yes	Yes
Fixed area	Yes	Yes	Yes	Yes	Yes
Fixed time	Yes	Yes	Yes	Yes	Yes
ρ	0.359 *	−0.345 *	−0.351 *	−0.279 *	−0.358 *
	(0.198)	(0.200)	(0.200)	(0.151)	(0.204)
Number of samples	360	360	360	360	360
R2	0.006	0.262	0.005	0.107	0.003
Log-L	−194.6941	312.7744	−192.4427	−280.0657	−187.1013

Note: (1) ***, **, and * represent significance levels of 1%, 5%, and 10%, respectively; (2) the values in parentheses in Models (1)–(5) are standard errors. Own compilation.

displays the estimated outcomes.

Firstly, carbon intensity can affect economic resilience by increasing the degree of industrialization. Table 6 illustrates the estimation results obtained from models (2) and (3), which somewhat indicate that the overall effect of carbon intensity on economic resilience is negative, but not significant, with a significant positive effect on the degree of industrialization. The degree of industrialization, the mediating variable, has a significant negative effect on economic resilience, implying that the degree of industrialization has a mediating effect on the process of carbon intensity affecting economic resilience. The bootstrap statistic for further testing was −1.654, which passed the significance test at the 1% level, indicating that the test results with mediating effects were robust. Due to the characteristics of resource endowments, fossil energy is the main energy supply in China. Therefore, fossil energy consumption, especially coal consumption, is the main source of carbon emissions. The increase in carbon intensity inevitably promotes economic growth, further increases the proportion of industries, and enhances the degree of industrialization [54]. The increase in carbon intensity has solidified the pollution-intensive industrial production mode, which is not conducive to the transition to a resource-saving production mode. Therefore, this result verifies the mediating effect of the industrialization degree on carbon intensity and economic resilience, that is, carbon intensity is not conducive to the transformation of production methods and thus hinders the improvement of economic resilience. This result validates Hypothesis 2 presented in this paper.

Secondly, another important mechanism by which carbon intensity affects economic resilience is the degree of pollution. The estimation results of models (4) and (5) in Table 6 show that the overall effect of carbon intensity on economic resilience is negative and significant, with a significant positive impact on pollution levels. Pollution level, the mediating variable, has a significant negative effect on economic resilience, implying that the pollution level plays a mediating role in the process of carbon intensity affecting economic resilience. The bootstrap statistic for further testing was −0.256, which passed the significance test at the 10% level, indicating that the test results with mediating effects were robust. The higher the carbon intensity, the more severe the environmental pollution. In recent years, China’s economic development has relied on industrialization, and energy consumption is closely related to pollutant emissions. Environmental problems are becoming increasingly serious, and the cost of pollution regulation and governance is increasing. The corresponding rise in the intensity of carbon emissions contradicts the concept of ecologically benign enhancement, which is incompatible with high-quality economic development [55]. Therefore, the pollution level is effectively verified, showing that reducing the carbon intensity level can reduce the environmental pollution and thus improve economic resilience. This result validates Hypothesis 3 presented in this paper.

6. Further Analysis

According to the abovementioned analysis, an increase in the carbon intensity level inhibits the improvement of economic resilience in the local and surrounding areas. At present, China’s energy consumption structure is mainly based on fossil fuels. Under the pressure of limited resources and ecological environment, it is an inevitable choice to effectively improve energy utilization efficiency and reduce carbon intensity levels. China has adopted a series of environmental regulatory policies, gradually exploring trading forms, such as emissions trading systems, carbon emissions rights, and energy rights. Among them, the carbon emissions permit trading policy (hereafter referred to as the carbon emissions trading policy) is a major institutional innovation that utilizes market mechanisms to control and reduce greenhouse gas emissions and promote green and low-carbon developments. The carbon emissions trading policy is not only conducive to guiding the optimization and upgrading of industrial structures, but can also drive the benign adjustment of the energy consumption structure, ensure sustainable development, and achieve high-quality economic development. To date, the pilot work of carbon emissions permit trading has been implemented for nearly a decade, and a unified, national carbon emissions trading market has been formed. However, can this pilot carbon emissions trading policy promote the improvement of economic resilience? What is the mechanism of action? This study regarded the carbon emissions permit trading pilot as a quasi-natural experiment and further analyzed the policy effectiveness and mechanism of the carbon emissions trading pilot on economic resilience using the DID model.

The development starting point of China’s carbon emissions trading market originated from the Notice on Pilot Work of Carbon Emissions Permit Trading issued by the General Office of the National Development and Reform Commission in October 2011. Since 2013, carbon emissions permit trading pilot projects have been launched in seven provinces and cities: Beijing, Tianjin, Shanghai, Chongqing, Guangdong, Hubei, and Shenzhen. Taking into account the availability of the data, this study utilized the panel data from 30 Chinese provinces and cities from 2010 to 2021 as the subjects of the study. The pilot provinces and cities mentioned above, including Beijing, Tianjin, Shanghai, Chongqing, Guangdong (including Shenzhen), and Hubei, were the experimental group, while the other provinces and cities were the control group. In the division of the pilot period, the officially implemented period from 2013 to 2021 was designated as the pilot period and the period from 2010 to 2012 was designated as the non-pilot period. This study selected the commonly used policy evaluation model, the DID model, to study the impacts of carbon emissions trading pilot projects on economic resilience. The benchmark regression model is set as follows:

$$Re{s}_{it}={\beta }_0+{\beta }_{1}trea{t}_i\times tim{e}_t+{\beta }_2{X}_{it}+{\mu }_i+{\gamma }_t+{\epsilon }_{it}$$

(3)

$Re{s}_{it}$ is the predicted variable, which measures the economic resilience of region i at time t; $trea{t}_i$ is the province grouping variable; $tim{e}_t$ is the time grouping variable; $X_{it}$ is the controlled variable; ${\mu }_i$ is a regional fixed effect; ${\gamma }_t$ is a fixed time effect; and ${\epsilon }_{it}$ is a random error term. In this context, i and t represent provinces and years, respectively.

6.1. Impact of Carbon Emissions Permit Trading Policies on Economic Resilience

The DID regression results are shown in

Table 7. DID benchmark regression results.

Variable Type	Model (1)	Model (2)	Model (3)	Model (4)
Policy effects	6.572 *	−3.290	12.373 ***	8.852 *
	(3.858)	(3.253)	(4.435)	(5.062)
Inf (infrastructure)	0.911	−1.797	11.642 **	4.513 *
	(1.380)	(2.186)	(5.469)	(2.622)
Fix (fixed-assets investment)	−2.332 *	0.640	−1.629	11.693 *
	(1.313)	(2.229)	(2.901)	(6.250)
Urb (urbanization level)	1.854	2.619	1.082	1.881 *
	(2.150)	(1.675)	(2.143)	(1.044)
Fore (degree of foreign trade)	5.585 **	5.121	3.872 **	−4.026
	(2.190)	(4.283)	(1.810)	(4.658)
Constant	10.622	−4.104	38.841	−20.092
	(15.782)	(20.716)	(23.150)	(22.306)
Control area	No	Yes	No	Yes
Control time	No	No	Yes	Yes
Number of samples	360	360	360	360
R2	0.066	0.178	0.135	0.252

Note: (1) ***, **, and * represent significance levels of 1%, 5%, and 10%, respectively; (2) the values in parentheses in models (1)–(4) are standard errors. Own compilation.

. Among them, model (1) is the regression result without controlling for time and region, and the regression coefficient of carbon emissions permit trading is significantly positive. Model (2) only controls the regression results of regions, and the regression coefficient of carbon emissions permit trading is not significant. Model (3) shows a regression result that only controls time, and the regression coefficient of carbon emissions permit trading is significantly positive. Model (4) controls the regression results of time and region, and the regression coefficient of carbon emissions permit trading is significantly positive, indicating that carbon emissions permit trading pilot policies can significantly improve economic resilience.

6.2. Intermediary Mechanism Testing

The abovementioned analysis further demonstrates that the pilot policy for carbon emissions permit trading has a positive effect on improving economic resilience. Therefore, what is the transmission mechanism by which this policy affects economic resilience? Based on the theoretical analysis, this study discussed the impact mechanism of carbon emissions trading policies on economic resilience from the perspectives of technology investment intensity and technology manpower investment. This study used the same regression method as in Formula (3), where the predicted variables in models (2) and (4) were technology investment intensity and technology manpower investment, respectively. Models (3) and (5) added technology investment intensity and technology manpower investment as the explanatory variables on the basis of Formula (1). The core explanatory variables, policy effects, and controlled variables remain unchanged. The specific empirical test results are shown in

Table 8. Mediating-effect test.

Variable Type	Model (1)	Model (2)	Model (3)	Model (4)	Model (5)
		Itec (Input Intensity in Science and Technology)		Manp (Technological Manpower Investment)
Policy effects	8.852 *	0.042 *	7.382	0.535 *	6.832
	(5.062)	(0.022)	(5.748)	(0.313)	(4.788)
Itec (input intensity in science and technology)			35.287 ***
			(9.317)
Manp (technological manpower investment)					3.773 *
					(2.101)
Constant	−20.092	−0.252 ***	−11.184	9.124 ***	−54.519 *
	(22.306)	(0.060)	(21.821)	(0.474)	(27.594)
Controlled variable	Yes	Yes	Yes	Yes	Yes
Control area	Yes	Yes	Yes	Yes	Yes
Control time	Yes	Yes	Yes	Yes	Yes
Number of samples	360	360	360	360	360
R2	0.252	0.678	0.256	0.985	0.255

Note: (1) *** and * represent significance levels of 1%, and 10%, respectively; (2) the values in parentheses in models (1)–(4) are standard errors. Own compilation.

.

Model (1) in Table 8 shows the regression results without mediating variables, with a policy effect coefficient of 8.852. Models (2) and (3) show a mediating effect on the intensity of technology investments. The estimated coefficient of carbon emissions permit trading on the intensity of technology investment is significantly positive, indicating that the implementation of carbon emissions permit trading pilot policies can strengthen the intensity of technology investments in the region. The regression coefficient of technology investment intensity in the region on economic resilience is significantly positive, indicating that the mediating effect of technology investment intensity should exist. Models (4) and (5) show the mediating effect of technological manpower investments, indicating that the carbon emissions permit trading pilot policy can promote regional technological manpower investments, and the regression coefficient of regional technological manpower investments on economic resilience is significantly positive, indicating that the mediating effect of technological manpower investment should exist. This result validates Hypotheses 4 and 5 presented in this study.

6.3. Analysis of Spatial Spillover Effects

The previous section examined the direct impact and underlying mechanisms of carbon emissions trading pilot policies on economic resilience, but pilot policies often presented indirect impacts on the economic resilience of neighboring regions due to the “leader” effect and radiating driving effect, namely spatial spillover effects. Therefore, based on the spatial correlation between regions and referring to the research methods of Du et al. [56], this study adopted the econometric method of spatial difference to test the spatial spillover effects of carbon emissions trading pilot policies and analyzed the spatial impact of policy implementations. This study used the spatial Durbin model’s DID method (SDM-DID) to explore the spatial spillover effects of carbon emissions trading pilot policies. Therefore, the SDM-DID model is as follows:

$$Re{s}_{it}={\lambda }_0+\rho WRe{s}_{it}+{\lambda }_{1}trea{t}_i\times tim{e}_t+{\lambda }_{2}W\left(trea{t}_i\times tim{e}_t\right)+{\lambda }_3{X}_{it}+{\mu }_i+{\gamma }_t+{\epsilon }_{it}$$

(4)

Among them, W is the spatial weight matrix, and the spatial distance matrix is selected to measure the connections between regions. The other variables are consistent with the public announcement (3).

The previous section presents the tested economic resilience through the Moran index, and there is a significant spatial correlation between the economic resilience levels on all regions from 2010 to 2021.

Table 9. Estimation results for the SDM-DID model.

Variable Type	Model (1)	Model (2)	Model (3)
	Main	Wx	Direct Effect	Indirect Effect	Total Effect
Policy effects	9.515 *	21.292 *	9.843 *	22.882 *	32.725 *
	(5.569)	(11.778)	(5.753)	(13.376)	(16.997)
Controlled variable	Yes	Yes	Yes
Fixed area	Yes	Yes	Yes
Fixed time	Yes	Yes	Yes
ρ	0.122 *
	(0.069)
Number of samples	360	360	360	360	360
R2	0.001	0.002	0.006	0.006	0.006
Log-L	−170.0959

Note: (1) * represent significance levels of 10%, respectively; (2) the values in parentheses in models (1)–(4) are standard errors. Own compilation.

presents the results of the spatial spillover effects based on the SDM-DID estimation. The estimated coefficient of policy effects in Model (1) in Table 9 is significantly positive, indicating that the carbon emissions trading pilot policy has effectively improved the economic resilience of the pilot areas, and confirms the benchmark regression results in Table 7. The direct effects and the DID estimation results in Model (3) are significantly positive. The results of Model (2) and indirect effects are significantly positive, indicating that the carbon emissions trading pilot policy has a positive spatial spillover effect, which is beneficial for improving the economic resilience of neighboring regions. The possible reason for this is that, with the implementation of carbon emissions trading pilot policies in pilot areas, the development model of the green economy emerges, and the introduction of relevant policies stimulates industrial transformations and upgrades. The “role model” and “leader” effects on pilot areas can effectively drive the development of green economies in neighboring areas and enhance the economic resilience of neighboring provinces in pilot areas [34]. Meanwhile, pilot areas tend to improve their carbon emissions through technological innovations, increasing the likelihood of neighboring areas being more quickly exposed to and introducing new technologies and learning new development models, thereby enhancing the economic resilience of neighboring areas [57,58].

7. Research Conclusions and Countermeasure Suggestions

7.1. Conclusions

Except for Tibet, this study was based on the provincial panel data obtained from 2010 to 2021 for 30 provinces in Mainland China. The spatial Durbin model was used to explore the spatial effect of carbon intensity on economic resilience. In a further analysis, the DID model was used to analyze the impacts and mechanisms of carbon emissions trading policies. Through the research and analyses of Hypotheses 1–5 in this study, and the following conclusions were drawn: (1) Both economic resilience and carbon intensity exhibited significant spatial heterogeneity results. The carbon intensity levels in Chinese provinces were distributed in a pattern of “high in the north and low in the south”, and the carbon intensity of the vast majority of provinces showed a trend of transition from high to low levels. From 2010 to 2021, the economic resilience of various provinces in China generally improved, with Guangdong, Shanghai, Zhejiang, Hubei, and other regions in high-value areas, while the economic resilience of northeast China showed a downward trend. (2) The local impact of carbon intensity on economic resilience was significant. The higher the intensity of carbon emissions in the local area, the more challenging it was to promote economic resilience. As the geographical distance increases, the intensity of carbon emissions can potentially impair the economic resilience of nearby regions. (3) The degree of industrialization and pollution played a mediating effect between carbon intensity and economic resilience. Carbon intensity was not favorable to the transition of production methods and consequently functioned as a hindrance to economic resilience. Reducing the carbon intensity can reduce environmental pollution and thus improve economic resilience outcomes. (4) The carbon emissions permit trading pilot policy can significantly enhance the economic resilience outcomes of the local and neighboring areas, and the carbon emissions permit trading pilot policy can promote the intensity of regional scientific and technological investments and scientific and technological manpower investments, thereby improving the economic resilience results.

7.2. Policy Implications

Based on the research conclusions of this study, the following policy implications were obtained:

Firstly, it is necessary to promote the coordinated development of emissions and carbon reductions in the north and south regions. During the sample period, there were significant differences in the carbon intensity levels between the southern and northern regions, which could be related to the gradual shift of regional differences in China’s economic development from being dominated by east–west differences to being dominated by the coexistence of east–west and north–south differences. Therefore, while maintaining the steady progress of emissions and carbon reductions in the southern region, greater efforts are needed to promote the acceleration of carbon intensity reductions in the northern region. On the one hand, it is necessary to provide the northern region integrated advantages in industrial structure adjustments, clean energy developments, and low-carbon transportation system constructions, so that policies can be tried first, gradually, and accelerate the process of carbon peaking and carbon neutrality in the northern region. On the other hand, it is recommended to further accelerate the construction of a national carbon emissions permit trading market, improve the allocation mechanism of carbon emissions quotas, incorporate more key industries into the trading system, and promote low-carbon coordinated developments in the north and south regions through market-oriented means.

Secondly, it is recommended to promote industrial structure upgrading, reduce carbon intensity levels, and achieve economic resilience improvements. In the process of China’s economic development, the adjustment of the industrial structure lacks a certain degree of rationalization and does not consider the negative effects of industrial developments under resource and environmental constraints. In the new period of high-quality economic developments in pursuit of economic resilience, it is critical to fully consider avoiding the excessive development of high-energy-consuming and high-polluting industries through industrial structure upgrades, and to guide the high-end upgrading and internal rationalization of industrial structures. Therefore, China’s economic development must promote the upgrading of industrial structures to achieve the goal of reducing carbon intensity levels and promote the energy low-carbon revolution and industrial low-carbon transformation, in order to achieve economic resilience improvements.

Thirdly, it is recommended to build a clean and low-carbon energy system to effectively reduce carbon and pollution emissions. China’s current energy structure is dominated by coal, and the proportion of primary electricity and other energy consumption sources is relatively low. China should prioritize investments in the development of new energy technologies, increase the development and utilization of new energy sources, such as hydrogen, and increase investments in renewable energy to improve energy efficiency levels and reduce carbon and pollution emissions. Meanwhile, the government should strengthen its systematic planning for low-carbon energy transformation; promote the process of low-carbon transformations from multiple aspects; establish technology-led, policy-driven, and market-driven energy transformation paths; improve the use of clean, low-carbon, and diversified energy systems; and provide development space for improving economic resilience by effectively reducing carbon and pollution emissions.

Fourthly, it is recommended to improve supporting measures for technological innovations and promote economic resilience. China’s low-carbon technology is still in its early stages of development, and the number of innovations is relatively low. In addition, the innovation of low-carbon technology involves co-ordination and cooperation in multiple fields, at various levels, and in various provinces. Therefore, it is necessary to strengthen the top-level design of the system, establish a low-carbon technology supply system that confronts different market demands in each province, and improve the incentive mechanism for technological innovation. Provinces and cities mainly focusing on pilot areas for carbon emissions permit trading should adapt to the local conditions, create a good cultural environment, simplify talent introduction policies, provide preferential policies for enterprise talent introductions, and promote the gathering of regional scientific and technological talents. It is necessary to increase the investments in science and technology, direct regional technological innovation, and incentivize local companies to come up with environmentally friendly innovations. Meanwhile, by fully utilizing interprovincial collaborative development and industrial integration strategies, an effective connection between the upstream, midstream, and downstream industrial technology systems should be formed according to the local conditions. A low-carbon technology research and development and achievement transformation development pattern should be achieved with government leadership, market promotion, and the coordination between the two. Technological innovation should be used to promote the development of a low-carbon economy, thereby effectively improving economic resilience outcomes.

7.3. Limitations and Future Research Directions

Two limitations of this study are proposed here. On the one hand, due to objective reasons, the data in this study could not be replaced with city-level data. The China Energy Statistical Yearbook is a comprehensive data collection source compiled by the Energy Statistics Department of the National Bureau of Statistics, which reflects China’s energy construction, production, consumption, and supply–demand balance values from 2008. It only includes provincial-level data. Meanwhile, the data required for the IPCC method are also missing from the city yearbook. Therefore, this study selected panel data samples at the provincial level. On the other hand, due to the limitations of the information platforms and investigation techniques, the collected information was not comprehensive enough and the data needed to be further enriched. In the future, the empirical research needs to be improved. With the increasing implementation of relevant policies, based on more detailed sample data, we will further verify and predict the influencing factors and internal theoretical mechanisms of economic resilience. Therefore, measures are planned to be proposed to improve economic resilience, enhance carbon emissions efficiency, and optimize carbon emissions-related policies. Meanwhile, members of various disciplines joined the research process, and the interdisciplinary communication and analysis from multiple perspectives helped to strengthen our problem-solving abilities.