Examining the Impact of Market Segmentation on Carbon Emission Intensity in China

Abstract

Market segmentation behavior has led to environmental issues, notably, carbon emissions, during China’s economic transformation, posing a significant challenge to ongoing economic progress. This study builds upon the existing literature and empirical analyses on market segmentation and carbon emissions, establishing a theoretical framework to understand the relationship between market segmentation and carbon emission intensity. Using a panel dataset covering Chinese provinces from 2003 to 2020, we systematically examined the impact and transmission mechanisms of market segmentation on carbon emission intensity. This study revealed the following findings: (1) there is a significant positive correlation between market segmentation and carbon emission intensity, which remains true after a series of robustness tests; (2) the promoting effect of market segmentation on carbon emission intensity in the eastern region is significantly lower than that in the central and western regions and, as time goes by, the promoting effect of market segmentation on carbon emission intensity becomes smaller and smaller; (3) resource mismatch and industrial structure are two important transmission channels through which market segmentation affects carbon emission intensity; (4) market segmentation demonstrated a distinct positive spatial contagion impact on carbon emission intensity. This study provides valuable policy recommendations to align with China’s ‘dual carbon goals’ by supporting a unified market and fostering low-carbon development. These recommendations aim to steer China’s economy toward a greener and low-carbon transformation.

1. Introduction

Climate change concerns, which include phenomena such as global warming, extreme disasters, and ecological imbalances triggered by carbon discharges, pose a significant threat to the socioeconomic advancement of humanity. This issue represents a common challenge for humanity in the 21st century. Therefore, all nations should prioritize the reduction of carbon emission intensity to promote environmentally friendly and sustainable low-carbon development [1,2,3]. With the rapid development of China’s economy, energy consumption and carbon emissions continue to increase. As a result, China has become the world’s largest primary energy consumer and carbon dioxide emitter [4,5,6]. According to the latest data, China’s carbon dioxide (CO₂) emissions were expected to reach approximately 11 billion tons in 2022, approximately constituting 28.87% of the world’s emissions and ranking it as the world’s top emitter. Thus, China’s CO₂ emissions have garnered international attention, intensifying the Chinese government’s urgency to reduce them [7,8].

As a responsible major country, China has pledged to strive to peak carbon dioxide emissions before 2030 and to achieve carbon neutrality by 2060. China has formulated important development strategies based on the concept of a community with a shared future for mankind. These strategies include ambitious goals aimed at achieving the “dual carbon targets,” which involve reducing carbon emission intensity and promoting green development [9]. The Chinese government is actively responding to global carbon reduction initiatives and has undertaken substantial efforts in this regard. The Chinese government emphasizes addressing carbon emissions and has issued numerous documents dedicated to this cause. China’s “14th Five-Year Plan” and the outline of long-term goals for 2035 set the target of “carbon dioxide emissions per unit of GDP in 2025 reduced by 18% compared with 2020.” These documents primarily focus on implementing the principles of low-carbon development, establishing an ecological civilization, and actively and steadily advancing carbon neutrality to accelerate China’s sustainable development. These endeavors hold immense practical significance in China’s pursuit of the “dual carbon targets.”

As an emerging economy in the world, China faces an important task in the current era: achieving a win–win situation between economic development and carbon emissions to reach the “double carbon” goal. In the process of China’s economic transformation and institutional reform, market segmentation remains a significant issue [10]. Market segmentation is a rational choice for cooperation and competition among local governments; however, it hinders the free and smooth flow of resources, factors, and technology across regions. Market segmentation is achieved through various measures that essentially “cut into pieces, shred, and mince” regional markets. This significantly hinders the flow of resources and industries across regions and restricts the coordinated development of regional economies [11]. Moreover, market fragmentation may further exacerbate environmental pollution and lead to high carbon emission intensity; however, this perspective remains controversial. On the one hand, market segmentation has intensified barriers between regions, impeding the free flow of resources, technology, labor, and knowledge factors. This has, to a certain extent, heightened energy consumption and contributed to an increase in carbon emission intensity [12,13]. On the other hand, market segmentation may motivate local governments to engage in a “race to the top” [14]. This involves strengthening the supervision of regional high pollution [15], curbing regional environmental pollution and carbon emissions. Clearly, the controversial research conclusions regarding the internal logical relationship between market segmentation and carbon emission intensity provide new opportunities for this article’s research. Therefore, under the new development pattern, studying the internal logical relationship between market segmentation and carbon emission intensity is crucial for China to promote carbon emission reduction and achieve the “dual carbon goals”.

While there is a growing body of research on carbon emissions and market segmentation, this field still requires further expansion. Firstly, current research predominantly focuses on the influencing factors and economic consequences of carbon emissions and market segmentation. There are few studies on carbon emission intensity related to market segmentation and further analysis and research are necessary. Secondly, the existing research predominantly centers on the correlation between market segmentation and the environment, largely overlooking the mechanics of market segmentation and its effect on carbon emission intensity. Thirdly, prior studies have neglected the spatial spillover effect that occurs between market segmentation and carbon emission intensity. Given that market segmentation affects the free flow of resources and factors between regions, it warrants thorough examination from a spatial perspective. Consequently, this study employs both empirical and theoretical analyses to deeply investigate the internal logical relationship between market segmentation and carbon emission intensity, utilizing the fixed effect model, intermediary effect model, and spatial effect model. Within the framework of the new development pattern, this research holds particular significance for China’s pursuit of the “dual carbon goals.”

Nevertheless, the current research is still in its initial stages of analyzing the relationship between market segmentation and carbon emission intensity. Compared with existing studies, this study focuses on three main potential contributors. First, it delves into the institutional factors that impact carbon emission intensity, thereby enriching the research landscape and offering a practical pathway for China to achieve its “dual carbon goals.” Second, this study comprehensively analyzes the impact of market segmentation on carbon emission intensity, examining the roles of resource mismatch and the upgrading of industrial structures in this process. Finally, this study explores the spatial spillover effects of market segmentation on carbon emission intensity. Through the exploratory research outlined above, this study provides theoretical support and guidance for decision-making to facilitate China’s effective transition to a low-carbon-development country.

This paper is organized into five main sections. Section 2 reviews the relevant literature. Section 3 describes the research hypotheses. Section 4 analyzes the empirical model, indicator design, and data sources. Section 5 comprehensively analyzes the empirical results. Section 6 summarizes the findings and provides policy recommendations.

2. Literature Review

During the era of rapid global development, there has been a substantial surge in energy consumption, leading to an increased focus on environmental concerns, particularly carbon emission intensity. China’s rapid economic growth has been closely associated with a development model characterized by short cycles and high rates of return [16]. However, this model inherently involves traditional industries with significant energy consumption and pollution levels, resulting in substantial carbon emissions. The ecological and environmental challenges stemming from this scenario create formidable barriers to achieving high-quality economic and social development. A reduction in carbon emission intensity can decrease the frequency of extreme weather events, mitigate the impacts of climate change, and contribute to improved environmental governance [17]. Clearly, reducing carbon emission intensity can promote global green and low-carbon development, facilitating a harmonious balance between economic growth and environmental preservation. Given this context, methods to tackle this issue have garnered widespread attention across various sectors of society.

Domestic and foreign scholars have conducted in-depth analyses and research on carbon emission intensity. Their findings have revealed that during China’s rapid development, economic growth, technological progress, scientific innovation, and industrial development have all influenced carbon emission intensity. Grossman and Krueger (1995) extensively explored the relationship between the economy and the environment, confirming the environmental Kuznets curve (EKC) theory [18]. Existing research indicates that the industrial structure has a very significant inhibitory effect on carbon emission intensity [19,20]. Zhu (2022) employed an econometric model to empirically analyze the impact of industrial structure on carbon emissions [21]. The author observed a significant negative relation between industrial structure and carbon emissions. In other words, industrial structure reduces carbon emissions. Furthermore, existing studies suggest that technological advancements can effectively reduce carbon emission intensity [22,23]. Wang et al. (2021) argued that foreign direct investment has a significant positive impact on carbon emission intensity. The authors conducted empirical tests to validate the pollution haven hypothesis [24]. The acceleration of urbanization has intensified energy consumption, which, in turn, has led to a significant increase in carbon emission intensity [25].

Furthermore, political factors, such as local government competition [26] and fiscal decentralization [27], influence carbon emission intensity. China has experienced significant market segmentation [28,29]. Scholars worldwide have explored the prominence of market segmentation amidst China’s economic transformation and institutional change, focusing on intentional restrictions on resource flow, factor mobility, and product exchange among regions through administrative interventions by local governments [30].

An increasing number of scholars are shifting their focus from the impact of market segmentation on economic growth to its impact on carbon emissions; two distinct views have emerged. The first perspective is the promotion theory, which suggests that market segmentation significantly intensifies carbon emission intensity [31,32]. As market segmentation intensifies, the utilization of resources within regions decreases, making it challenging to harness comparative advantages and leading to a deterioration in environmental quality, which, in turn, impacts carbon emission intensity. The market segmentation strategy of “taking advantage of neighbors” results in economic isolation between regions, affecting resource flow and economic exchanges between them. This restriction on coordinated and cooperative development leads to resource wastage and exacerbates carbon emission intensity. The second perspective is the suppression theory, suggesting that market segmentation suppresses carbon emission intensity [33,34]. The central government’s performance assessment model has sparked a “race to the top” among local governments, encouraging them to enhance environmental standards and reduce carbon emission intensity.

3. Research Hypotheses

3.1. The Impact of Market Segmentation on Carbon Emission Intensity

Market segmentation is a typical phenomenon in China’s development process [35]. It represents a rational choice for local governments looking to expedite economic development and gain political advantages, resulting in intense competition. First of all, the overall efficiency of the market segmentation market has declined, resulting in the inability to achieve optimal allocation of resources [36]. As the degree of market segmentation deepens, the economic ties between regions are severed, leading to the distortion of the economic operating mechanism, which is not conducive to the optimal allocation of resources, which, in turn, leads to higher carbon emission intensity. Secondly, market segmentation inevitably weakens regional competitiveness, restricts the formation of regional competition mechanisms, and weakens the internal competitive motivation of enterprises for large-scale production and green technology innovation and development. It reduces the overall regional production efficiency and hinders the development of green innovation, thereby exacerbating regional carbon emissions and causing carbon emission intensity to remain high. In summary, this paper presents Research Hypothesis 1:

Hypothesis 1.

Market segmentation has a positive effect on carbon emission intensity.

3.2. The Impact Mechanism of Market Segmentation on Carbon Emission Intensity

For an extended period, local governments have steadfastly prioritized regional economic development in their pursuit of economic growth. This emphasis, mandated by the central government’s primary assessment criteria, has fueled intense competition for resources and markets, ultimately exacerbating market fragmentation. Market segmentation results in inefficient resource utilization, leading to market disarray and hindering effective socioeconomic development [37]. This, in turn, exacerbates resource mismatch and distortion phenomena. Market segmentation hinders the high-level free flow of technology, labor, and other resources; reduces the efficiency of resource allocation; and becomes an important transmission mechanism for market segmentation to affect carbon emission intensity. First, intense competition among local governments worsens market segmentation, resulting in significant fragmentation between regions. Consequently, there is inefficient resource allocation and reduced energy efficiency, further exacerbating carbon emission intensity [38]. Moreover, market fragmentation resulting from competition among local governments creates administrative barriers that partition regional markets and obstruct the flow of factors, resources, and economic connections between regions. These barriers hinder the transfer of regional resource endowments to areas with higher marginal productivity, leading to regional imbalances and decreased efficiency. Consequently, this amplifies the impact of carbon emission intensity, intensifying its magnitude. Accordingly, Hypothesis 2 is proposed as follows:

Hypothesis 2.

Market segmentation affects carbon emission intensity by affecting resource misallocation.

Market segmentation significantly hinders the adjustment of the industrial structure and is not conducive to the process of industrial structure adjustment [39], becoming an important transmission mechanism for market segmentation to affect environmental pollution. Under the influence of local protectionism and promotion tournaments, local governments establish trade barriers, monopolize the supply of factors, or impede the introduction of new technologies to serve the interests of individuals or groups [40]. This has led to the concentration of high-tax, high-pollution, traditional low-end industries within the region is observed and the slow upgrading of the industrial structure. These industries are still the main source of carbon emissions [41]. Market segmentation, resulting from competition among local authorities, exacerbates regional barriers. This prompts the “imitation of high-profit industries” and promotes the homogenization and low-end development of regional industries. Consequently, this situation leads to resource wastage, increased energy consumption, and intensified carbon emissions. Driven by the promotion mechanisms and the pursuit of economic development, local governments find themselves in the “low-end trap” concerning their industrial development. This scenario hinders the adoption of clean technologies and the development of low-carbon industries. Accordingly, Hypothesis 3 is formulated as follows:

Hypothesis 3.

Market segmentation influences carbon emission intensity through its impact on the industrial structure.

3.3. Spatial Spillover Effects of Market Segmentation on Carbon Emission Intensity

Market segmentation is a unique phenomenon in China’s economic transformation and development process. The “beggar-thy-neighbor” market segmentation strategy has intensified competition among local governments. To promote local economic development and serve their own interests, local governments restrict the cross-regional flow and dissemination of resources, factors, and products, impeding the breadth and depth of exchanges. At the same time, market segmentation has spatial spillover effects on economic growth [42], reduces energy efficiency [43], and aggravates environmental pollution [44]. The impact of market segmentation on economic growth comes from the spatial connections between different regions; the corresponding impact of market segmentation on carbon emission intensity will also have spatial effects. Market segmentation hinders the interconnection between regions and has a certain impact on adjacent areas, which, in turn, produces spatial effects. Therefore, it is necessary to establish a spatial econometric model to examine the spatial spillover effect of market segmentation on carbon emission intensity. Consequently, this article presents the following hypotheses:

Hypothesis 4.

Market segmentation has a spatial spillover effect on carbon emission intensity.

4. Model, Data, and Variables

4.1. Model Specification

We constructed a benchmark model to examine the inherent relationship and the mechanism of influence between market segmentation and carbon emission intensity, as shown in Equation (1). Additionally, we applied logarithmic transformations to the relevant variables to address heteroskedasticity:

$${CEI}_{it}={\alpha }_0+{{\alpha }_{1}lnMS}_{it}+{\alpha }_n{\sum }_{d=2}^p{X}_{it}+{\lambda }_i+{\mu }_t+{\epsilon }_{it}$$

(1)

where CEI denotes total carbon emission intensity and MS denotes market segmentation. Furthermore, we introduced control variables X, including human capital (HR), environmental regulation (ER), transport infrastructure (TI), and population density per capita (PD), to prevent the omission of important explanatory variables from generating endogeneity problems. Moreover, i and t denote provincial (municipal) regions and years, respectively; ${\mathsf{\alpha }}_0$ is a constant term; ${\mathsf{\alpha }}_1$ denotes the coefficient corresponding to the explanatory variable of market segmentation; and ${\mathsf{\alpha }}_{\mathrm{n}}$denotes the coefficient corresponding to the control variable. Additionally, ${\mathsf{\lambda }}_{\mathrm{i}}$ represents the individual effect, ${\mu }_t$ represents the time effect, and ${\epsilon }_{it}$represents the random interference term.

We used the mediating effect model to construct Models (2) and (3) to assess the role of resource mismatch in mediating the relation between market segmentation and carbon emission intensity, as proposed in Hypothesis 2:

$${RM}_{it}={\beta }_0+{{\beta }_{1}lnMS}_{it}+{\beta }_n{\sum }_{d=2}^p{X}_{it}+{\epsilon }_{it}$$

(2)

$${CEI}_{it}={\gamma }_0+{{\gamma }_{1}lnMS}_{it}+{{\gamma }_{2}lnRM}_{it}+{\gamma }_n{\sum }_{d=3}^p{X}_{it}+{\epsilon }_{it}$$

(3)

where RM denotes a proxy variable for regional resource mismatch; the other variables have been explained earlier. Testing the intermediate effect model involves examining the significance of the coefficient β1 in Model (2) when the coefficient of ${\alpha }_1$ in Model (1) is positive and significant. We assessed the effects of market segmentation and resource mismatch on carbon emission intensity, with coefficients denoted as ${\gamma }_1$ and ${\gamma }_2$, respectively.

Furthermore, the mediating effect model was used to construct Models (4) and (5) to assess the mediating role of industrial upgrading between market segmentation and carbon emission intensity proposed in Hypothesis 3:

$${ST}_{it}={\delta }_0+{{\delta }_{1}lnST}_{it}+{\delta }_n{\sum }_{d=2}^p{X}_{it}+{\epsilon }_{it}$$

(4)

$${CEI}_{it}={\vartheta }_0+{{\vartheta }_{1}lnMS}_{it}+{{\vartheta }_{2}lnST}_{it}+{\vartheta }_n{\sum }_{d=3}^p{X}_{it}+{\epsilon }_{it}$$

(5)

where ST denotes a proxy variable for regional industrial upgrading; the other variables are explained above.

4.2. Description of Variables

(1)

Explanatory Variables

Our primary focus was on carbon emission intensity (CEI) as the core explanatory variable. With the progression of global industrialization and urbanization, worldwide energy consumption, particularly of traditional fossil fuels, has been on the rise. Consequently, total CO₂ emissions have increased. To assess China’s actual carbon emission status, we adopted a reference method proposed by the Intergovernmental Panel on Climate Change in 2006. We selected eight fossil fuels closely related to carbon emissions, including raw coal, coke, crude oil, gasoline, kerosene, diesel, fuel oil, and natural gas, among others, to estimate total carbon emissions. This study measures carbon emission intensity as the ratio of total carbon emissions in provincial (municipal) regions to GDP, following previous studies by Zhang et al. [45]. The specific calculation steps were as follows: first, regional energy consumption was estimated and total regional carbon emissions were assessed based on individual energy carbon emission factors; second, carbon emission intensity was assessed using the ratio of total regional carbon emissions to GDP. The GDP at the provincial and regional levels used in our study does not involve data from Hong Kong, Macao, Taiwan, and Tibet due to the availability and continuity of data.

The specific calculation formula is as follows:

$$CEI=\frac{C{O}_2}{GDP}$$

(6)

(2)

Core Explanatory Variables

We primarily investigated market segmentation (MS), a form of local protectionism in which local governments employ various measures to restrict the free flow of resources within a region [46,47]. Measurement methods generally include production methods [48], price methods [49], and trade law [50]. The production method mainly analyzes the level of market segmentation by measuring the degree of identical composition of regional industries. Its disadvantage is that the calculation is relatively simple. Because the industry classification is relatively broad, it cannot truly and objectively reflect the level of regional market segmentation. Trade law mainly measures the level of regional market segmentation through inter-regional trade flows. Due to the availability and continuity of data, it cannot describe and analyze the level of regional market segmentation well. The price method mainly reflects the level of market segmentation based on the price differences of different industries in the region. It is relatively more advantageous and is, therefore, widely used. In the development process of our country, product marketization has developed rapidly and the market prices of products in different regions vary greatly. Therefore, the diversification of product prices within the region should be fully considered to reflect the market segmentation of different regions. In this study, we adopted the research approach of Lai et al. [51] and used the relative price method to measure market segmentation in China’s provinces. The calculation process is as follows: this model assumes the presence of transaction costs, leading to price fluctuations within a certain range between two locations. We constructed a panel data structure involving three dimensions: time (t), region (b, c), and commodity (k). These three-dimensional (t × b × k) panel data formed the foundation for our further analysis. Given the availability of comprehensive and continuous data, we selected eight product categories for analysis: food, apparel and footwear, beverages and tobacco, cultural and sporting goods, pharmaceuticals, printed materials, everyday essentials, and fuel.

First, we calculated the absolute values of the relative prices between regions. Since the data used were chain indices of retail commodity prices, the price differential can be expressed in the following format:

$$\mathsf{\Delta }{Q}_{bct}^k=\mathit{ln}\left(P{I}_{bt}^k\right)-\mathit{ln}\left(P{I}_{ct}^k\right)$$

(7)

This equation indicates that in a dataset comprising thirty provincial and municipal areas and data for eight categories of commodities, from 2003 to 2020 (18 years), there exist 435 pairs of provincial and municipal combinations, resulting in 62,640 absolute values of relative prices in differential form.

Second, we used a demeaning method to process the commodity price data for effectively eliminating systematic bias. Assuming that Δ${\mathrm{Q}}_{\mathrm{b}\mathrm{c}\mathrm{t}}^{\mathrm{k}}={\mathrm{a}}^{\mathrm{k}}+{\mathsf{\epsilon }}^{\mathrm{k}\mathrm{a}\mathrm{b}\mathrm{t}},$ and considering that ${\mathrm{a}}^{\mathrm{k}}$ is solely associated with the commodity type k and ${\mathsf{\epsilon }}^{\mathrm{k}\mathrm{b}\mathrm{c}\mathrm{t}}$ is closely linked to the market conditions of regions b and c, we calculated the average absolute relative price values for the 435 pairs of provincial and municipal areas for commodity k:

$$q_{bct}^k=\left|\mathsf{\Delta }{Q}_{bct}^k\right|-\left|\overline{\mathsf{\Delta }{Q}_t^k}\right|=\left(a^k-a^{-k}\right)+\left({\epsilon }_{bct}^k-{\epsilon }_{bct}^{-k}\right)$$

(8)

Following this procedure, we gathered information regarding inter-regional market segmentation and specific stochastic factors.

Finally, we calculated the data on the relative price variances between provincial and municipal regions for the 435 pairs by assessing the variance in relative price fluctuations for the eight categories of commodities in regions m and n during year t. The market segmentation index for each provincial and municipal area concerning the rest of the country was determined by aggregating the data at the provincial and municipal levels:

$$\mathit{Var}\left(q_{nt}^k\right)=\left({\sum }_{b\ne c}\mathit{Var}\left(q_{bct}^k\right)\right)N$$

(9)

In this context, n denotes the region and N denotes the total count of combined provincial and municipal portfolios. We collected 540 observations (30 × 18) after implementing the measurement process described previously.

(3)

Intermediary Variables

Based on the previous theoretical analysis, this paper selects market segmentation and industrial structure as intermediary variables to examine the intermediary transmission mechanism of market segmentation on carbon emission intensity. Market segmentation is not conducive to the cross-regional flow of labor, technology, capital, and other resource elements; reduces resource allocation efficiency; and, to a certain extent, intensifies carbon emission intensity. This paper analyzes resource misallocation as a mediating variable between market segmentation and carbon emission intensity. Drawing on the concept of labor mismatch, as defined by Du and Li [52], we measured the level of resource mismatch in various provinces, cities, and regions in China. Market fragmentation has severed the close ties between regions, restricted the coordinated development and scale of industries, inhibited the green upgrading of industries, and led to a continuous increase in carbon emission intensity.. Therefore, this paper analyzes industrial structure as an intermediary variable between market segmentation and carbon emission intensity. In addition, following the study by Yu et al. [53], we employed the quality of industrial restructuring as an indicator of industrial structure.

(4)

Control Variables

There are many factors affecting carbon emission intensity. This paper introduces a series of control variables. As the level of human capital increases, its role in promoting the development of green and low-carbon economies and reducing carbon emission intensity becomes more and more obvious; it is the main factor affecting carbon emission intensity. Human capital (HR) is measured by the number of years of education per capita in each province. Environmental supervision is an important means for the government to guide and manage ecological and environmental issues. Improving the level of environmental supervision has an important impact on reducing carbon emission intensity. As mentioned by Yang et al. [54], environmental regulation (ER) is measured by the ratio of pollution control investment to GDP. Transportation infrastructure shortens the time and space distance, helps optimize resource allocation, and improves energy utilization efficiency, thereby affecting carbon emission intensity. Transportation infrastructure (TI) is quantified as the ratio of population to regional area. It serves as an indicator of regional infrastructure development following the study by Zhang et al. [55]. The greater the regional population density, the greater the scale of economic development and energy consumption, which will inevitably have an impact on carbon emission intensity. Population density (PD) is determined by dividing the total population at the end of each year by the administrative area of each region, in accordance with the study by Song et al. [56].

4.3. Data Sources and Statistical Analyses

Taking data availability and authenticity into account, we conducted our analysis at the provincial level in China, excluding Hong Kong, Macao, Taiwan, and Tibet, resulting in 30 regions. The primary data sources for this study included the China Energy Statistics Yearbook (2003–2020), the China Statistical Yearbook (2003–2020), regional statistical yearbooks (2003–2020), and the Wind Database (2003–2020). In cases of missing data during the data collection process, we applied interpolation methods to fill the gaps.

Table 1. Descriptive statistics of variables.

Variables	Obs	Mean	Std.	Min	Max
CEI	540	3.2022	2.789	0.2167	19.915
lnMS	540	−8.5475	0.5967	−9.9356	−6.3302
lnRM	540	−1.2432	0.9960	−6.6322	1.2307
lnST	540	2.0700	0.6112	0.4265	3.4576
lnHR	540	2.1668	0.1175	1.7984	2.5480
lnER	540	−5.921	0.86756	−10.022	−2.2701
lnTI	540	−0.4789	0.84062	−3.381	0.7907
lnPD	540	5.4401	1.2781	2.0005	8.2812

presents the descriptive statistics for each variable.

5. Empirical Results and Discussion

Prior to the econometric model analysis, we conducted multiple cointegration and correlation tests, the specific results of which are presented in

Table 2. Variance inflation factors for key variables.

Variables	VIF	1/VIF
lnMS	1.19	0.84
lnHR	1.43	0.69
lnER	1.24	0.81
lnTI	1.32	0.76
lnPD	1.41	0.71

and

Table 3. Correlation coefficient analysis.

	CEI	lnMS	lnHR	lnER	lnTI	lnPD
CEI	1.000
lnMS	0.175	1.000
lnHR	−0.412	−0.176	1.000
lnER	0.527	0.066	−0.385	1.000
lnTI	−0.267	−0.363	0.104	−0.125	1.000
lnPD	−0.433	−0.027	0.441	−0.296	−0.154	1.000

. As indicated in Table 2, the maximum variance inflation factor (VIF) value was 1.41. The VIF values for all variables were <10, confirming the absence of multicollinearity among the main variables. Table 3 indicates a noticeable level of correlation among the main variables, implying the feasibility of subsequent model analysis and research.

5.1. Benchmark Regression Results

In order to deeply explore the mechanism of market segmentation on carbon emission intensity and ensure the reliability of the empirical results, this article uses the OLS, FE, and RE models for regression analysis. The results are presented in

Table 4. The estimated results based on the basic model.

Variables	Explained Variable: CEI
	(1)	(2)	(3)
	FE	RE	OLS
lnMS	0.247 ***	0.234 ***	0.102
	(0.064)	(0.065)	(0.094)
lnHR	−8.378 ***	−6.163 ***	−1.532 **
	(0.861)	(0.735)	(0.633)
lnER	0.198 ***	0.222 ***	0.745 ***
	(0.053)	(0.053)	(0.087)
lnTI	−0.668 ***	−0.966 ***	−1.166 ***
	(0.187)	(0.174)	(0.156)
lnPD	1.959 ***	−0.266 *	−0.447 ***
	(0.634)	(0.145)	(0.058)
Constant	15.166 ***	23.257 ***	16.692 ***
	(2.858)	(1.216)	(1.515)
Observations	540	540	540
Number of id	30	30	30
R-squared	0.586	0.575	0.443
Hausman (p-value)	0.000

Note. *, **, and *** indicate the levels of significance at 10%, 5%, and 1%, respectively, while () represents the standard error. The table below follows the same notation.

. Whether utilizing the fixed effects model, random effects model, or OLS regression model, the coefficient of the core explanatory variable market segmentation is significantly positive. This indicates a positive correlation between market segmentation and carbon emission intensity. The Hausman test results show that this article adopts the fixed effects model as the benchmark regression model. The results are shown in Column (1) of Table 4.

According to the benchmark regression results presented in Table 4, Column (1), it can be seen that the estimated coefficient of the core explanatory variable market segmentation is positive and statistically significant. This observation indicates that market segmentation significantly exacerbates carbon emission intensity. Firstly, local governments engage in fierce competition to expand their political and economic gains. This intensifies market fragmentation and reinforces regional barriers associated with “beggar-thy-neighbor” policies, disrupting regional development and leading to regional duplication, resource wastage, and increased carbon emissions. Consequently, this contributes to elevated levels of carbon emission intensity. Secondly, local governments compete for “growth,” driven by performance appraisal mechanisms that emphasize economic growth. This competition results in reduced ecological and environmental oversight, with a focus on industries possessing political appeal to achieve economic catch-up and capital accumulation. Industries with political appeal often exhibit high energy consumption, high emissions, and resource-intensive operations, thereby increasing regional carbon emission intensity. This finding aligns with Hypothesis 1.

With regard to the influence of control variables on carbon emission intensity, an inverse association was observed between human capital and carbon emission intensity. This finding indicates that the higher the levels of human capital in a region, the lower its carbon emission intensity. Thus, increases in human capital can substantially mitigate carbon emission intensity. Accordingly, regions with higher levels of human capital experience more pronounced knowledge and technological spillover effects. This situation, in turn, leads to higher adoption rates of environmentally friendly technologies and practices, thereby reducing carbon emission intensity. Furthermore, transport infrastructure was observed to have a significant inhibitory effect on carbon emission intensity. The construction and enhancement of transport infrastructure effectively shorten the distance between regions, thereby reducing transport costs, resource consumption, and carbon emission intensity. Considering that local governments have relaxed environmental regulations in pursuit of growth, environmental regulation and carbon emission intensity seem to be positively correlated. Perhaps, this positive correlation can be attributed to the fact that environmental regulation standards have been lowered by local governments in the pursuit of economic development. This situation has resulted in a constant inflow of resource-intensive and labor-intensive enterprises and has increased energy consumption and carbon emission intensity. Finally, the continuous increase in population density has increased the carbon emission intensity. Along with the urbanization process, the economic level and population continue to increase. Furthermore, areas with high population density exhibit inefficient land use, traffic congestion, and high energy consumption in buildings, leading to increased carbon emission intensity.

5.2. Robustness Analysis

We conducted a series of robustness tests to further strengthen the reliability of the results. These tests encompassed the exclusion of particular municipalities, reduction of the sample duration, and substitution of alternative measures for the control variables.

(1)

Excluding municipalities directly under the central government

Compared with other provinces, the four municipalities directly under the central government, namely, Beijing, Shanghai, Tianjin, and Chongqing, possess superior geographical and resource endowment advantages. We considered their higher economic development levels to minimize significant differences in market segmentation. Furthermore, we eliminated four municipalities directly under the central government and repeated the benchmark model regression analysis.

Table 5. Robustness test of the baseline regression.

Variables	Explained Variable: CEI
	(1)	(2)	(3)
	Delete Municipality	2003–2019	Substitution of Control Variables (TI)
lnMS	0.151 **	0.117 *	0.139 **
	(0.067)	(0.066)	(0.065)
lnHR	0.034 *	0.019	−7.833 ***
	(0.019)	(0.019)	(0.754)
lnER	0.320 ***	0.383 ***	0.159 ***
	(0.081)	(0.068)	(0.052)
lnTI	1.217 ***	1.712 ***	−0.130 **
	(0.321)	(0.312)	(0.057)
lnPD	1.519 ***	1.289 ***	−0.742 ***
	(0.157)	(0.139)	(0.149)
Constant	7.955 ***	6.483 ***	2.756 ***
	(0.770)	(0.733)	(0.587)
Observations	468	510	540
Number of id	26	30	30
R-squared	0.719	0.712	0.609

Note. *, **, and *** indicate the levels of significance at 10%, 5%, and 1%, respectively, while () represents the standard error.

presents the robust regression results for the removal of municipalities. The findings indicate that the coefficient of the core explanatory variable market segmentation was 0.151, which was significant at the 5% level. This finding implies that the increased market segmentation further increases carbon emission intensity, confirming the validity of the previous results and the robustness of the regression results.

(2)

Shortening the sample duration

Considering the impact of the COVID-19 epidemic on China, the data in 2020 may also be affected by fluctuations. To ensure the reliability and accuracy of the research results, we deleted the data in 2020 and re-conducted baseline regression analysis. The results of this regression analysis are presented in the second column of Table 5. Clearly, the market segmentation variable maintained a positive coefficient and passed the significance test, thereby substantiating the robustness of the regression findings. Any potential biases related to the time period were minimized by shortening the time span of the sample.

(3)

Measures of variation for control variables

We assessed the measure of transport infrastructure using the per capita ownership of private cars in each province as a substitute variable. The data were subjected to regression analysis. The results are displayed in Table 5. Market segmentation, the primary independent variable, exhibited a statistically significant positive relation. This finding emphasizes the stability and reliability of the initial regression results, confirming their credibility and validity. These findings suggest that market segmentation positively impacted carbon emission intensity.

In sum, the estimates obtained, whether by excluding municipalities, shortening the sample time, or substituting control variables, consistently aligned with the findings in Table 5. This suggests the validity of the previous hypotheses and conclusions, indicating relatively robust baseline regression results.

5.3. Heterogeneity Analysis

We examined temporal and spatial diversity to delve into the possible differences in the impact of market segmentation on carbon emission intensity.

(1)

Temporal heterogeneity

The Third Plenary Session of the 18th Party Central Committee in 2013 placed considerable emphasis on the establishment of an integrated and regulated market system in China. Moreover, a shift was observed in the evaluation criteria used by the central government to assess the performance of local governments, increasing the proportion of environmental protection. Under the influence of the promotion mechanism, local government officials continue to increase regional environmental protection and promote regional green and low-carbon transformation and development, thus having a greater impact on regional carbon emission intensity. Accordingly, we used 2013 as a reference point and investigated temporal disparities with regard to the impact of market segmentation on carbon emission intensity. We specifically examined two distinct time intervals: 2000–2012 and 2013–2020.

As depicted in

Table 6. Heterogeneity regression results.

Variables	Explained Variable: CEI
	Time Heterogeneity		Regional Heterogeneity
	2003–2012	2013–2020	Eastern Region	Central and Western Regions
lnMS	0.270 ***	−0.050	0.122 **	0.195 **
	(0.103)	(0.052)	(0.061)	(0.094)
lnHR	−5.532 ***	−4.314 ***	−4.888 ***	−6.936 ***
	(1.282)	(0.896)	(1.011)	(1.199)
lnER	0.322 ***	−0.015	0.059	0.222 ***
	(0.082)	(0.040)	(0.050)	(0.077)
lnTI	−0.534 **	−0.967 ***	−0.578 ***	−1.531 ***
	(0.241)	(0.249)	(0.169)	(0.307)
lnPD	0.596	−2.137 *	−2.436 ***	8.043 ***
	(1.117)	(1.121)	(0.658)	(1.217)
Constant	17.146 ***	25.338 ***	31.275 ***	−14.206 **
	(5.428)	(6.165)	(2.835)	(5.796)
Observations	300	240	198	342
Number of id	30	30	11	19
R-squared	0.380	0.379	0.680	0.626

Note. *, **, and *** indicate the levels of significance at 10%, 5%, and 1%, respectively, while () represents the standard error.

, market segmentation exhibited a positive correlation with carbon emission intensity during 2000–2012. Perhaps, the primary rationale is that market segmentation considerably impedes the unrestricted movement and allocation of resources, thereby undermining the efficient flow and allocation of resources between industries. To some extent, resource wastage results in high carbon emission intensity. Second, local governments prioritize economic development at the expense of environmental protection in pursuit of competitive growth. This situation leads to excessive investment in energy-intensive and high-capacity industries, as well as resource-intensive practices, ultimately exacerbating carbon emission intensity. During 2013–2020, a negative relation was observed between market segmentation and carbon emission intensity; however, the coefficient was not statistically significant. The construction of a unified market in China has gradually strengthened regional exchanges and cooperation but weakened the regional barrier of “beggar-thy-neighbor.” This situation has resulted in the free flow of regional factors of production and the effective allocation of resource factors. This, to a certain extent, has reduced carbon emission intensity. Furthermore, the continuous growth of China’s economy has increased public awareness regarding environmental protection, resulting in a higher demand for environmental quality. The central government is increasingly recognizing the significance of environmental preservation. The focus of government officials is shifting from a sole emphasis on “GDP” to “effective environmental management.” Additionally, the paradigm of “bottom–up competition” is evolving toward “top–down competition.” Accordingly, local governments are adopting stricter environmental standards, intensifying their environmental protection efforts and, thereby, effectively reducing carbon emission intensity.

(2)

Spatial heterogeneity

Given that regions vary in terms of economic development and policy formulation, differences may be observed in the resources invested in and the importance attached to market segmentation and carbon emission intensity. This regional variation can result in a regionally heterogeneous impact of market segmentation on carbon emission intensity. To address this inquiry, we divided the overall sample into two major regions: the east and the midwest. Furthermore, we conducted separate benchmark regressions. Table 6 presents the results of these regression analyses.

As illustrated in Table 6, the results of the regional heterogeneity analysis illustrated that market segmentation significantly contributes to carbon emission intensity in the eastern and central-western regions. The impact of market segmentation on carbon emission intensity varies across regions, with a more pronounced promotional effect observed in the central-western region compared to the eastern region. This difference may be attributed to the relatively less developed infrastructure and economic progress in the central-western region. In these areas, local governments ignore environmental issues in their excessive pursuit of economic development. Consequently, ecological and environmental regulatory standards may be compromised, leading to an increase in carbon emissions. In contrast, the eastern region is known for its advanced economic development and local governments with greater financial strength. These governments adopt a “top-by-top” competitive strategy, emphasizing investment in regional ecological and environmental management.

5.4. Transmission Mechanism Test

This study discusses the resource misallocation and industrial upgrading mechanisms of market segmentation that affect carbon emission intensity.

We estimated the mediating effect model represented by Models (2)–(5) to test Hypotheses 2 and 3 and examine the mechanisms of the effect of market segmentation on carbon emissions. The estimated results are presented in

Table 7. Results of the transmission mechanism test.

Variables	Benchmark	Resource Mismatch		Industrial Structure
	(1)	(2)	(3)	(4)	(5)
	CEI	RM	CEI	ST	CEI
lnMS	0.247 ***	0.090 *	0.232 ***	−1.383 ***	0.161 **
	(0.064)	(0.046)	(0.064)	(0.223)	(0.065)
lnRM			0.159 **
			(0.062)
ST					−0.062 ***
					(0.012)
lnHR	−8.378 ***	−0.025	−8.374 ***	14.060 ***	−7.513 ***
	(0.861)	(0.618)	(0.857)	(3.016)	(0.860)
lnER	0.198 ***	0.108 ***	0.181 ***	−0.250	0.183 ***
	(0.053)	(0.038)	(0.053)	(0.185)	(0.052)
lnTI	−0.668 ***	0.163	−0.694 ***	3.787 ***	−0.435 **
	(0.187)	(0.134)	(0.186)	(0.654)	(0.189)
lnPD	1.959 ***	−0.892 *	2.100 ***	26.209 ***	3.571 ***
	(0.634)	(0.455)	(0.633)	(2.221)	(0.700)
Constant	15.166 ***	4.653 **	14.428 ***	−186.612 ***	3.684
	(2.858)	(2.052)	(2.857)	(10.009)	(3.630)
Control	YES	YES	YES	YES	YES
Observations	540	540	540	540	540
Number of id	30	30	30	30	30
R-squared	0.586	0.050	0.592	0.667	0.605

Note. *, **, and *** indicate the levels of significance at 10%, 5%, and 1%, respectively, while () represents the standard error.

. Column (1) indicates the results of the baseline regression of the impact of market segmentation on carbon emission intensity. As indicated in Columns (2) and (3) of Table 7, market segmentation affected carbon emission intensity through resource misallocation. The stepwise analysis theory posits a significant mediating effect, i.e., market segmentation impedes the unrestricted movement of factors and distorts resource allocation, causing an increase in carbon emission intensity. This phenomenon may be attributed to intensified competition among local governments, further reinforcing market segmentation. This scenario leads to a relatively closed and fragmented market with clear barriers between regions, preventing resource factors from entering more efficient regions. Consequently, the ineffective utilization of regional resources and substantial resource inefficiency, to a certain degree, intensify carbon emissions. As implied in the regression results of the transmission mechanism in Columns (4) and (5) of Table 7, market segmentation exacerbated carbon emission intensity by hindering the industrial structure. This may be attributed to the fact that local governments invest substantial capital in traditional, high-energy, and polluting industries in their pursuit of economic development. This hinders the pace of regional green and low-carbon development, thereby exacerbating carbon emission intensity. Based on the above analysis, market segmentation impacts carbon emission intensity through its influence on resource misallocation and the industrial structure. This finding validates the hypotheses mentioned earlier, namely, Hypotheses 2 and 3.

5.5. Further Extension of the Analysis

We introduced spatial interaction terms based on the baseline econometric model and constructed a spatial model to examine the impact of market segmentation on carbon emission intensity and further investigate the spatial spillover effects of market segmentation on carbon emission intensity. The Moran test results show that there is a strong spatial correlation between market segmentation and carbon emission intensity. We employed a spatial adjacency weight matrix to analyze the spatial effects of market segmentation on carbon emission intensity. Prior to conducting the empirical analysis, it is essential to perform relevant tests, such as the LM test and LR test, on the econometric model to clarify the correctness and rationality of the model selection. We draw on the research ideas of Chen et al. [57] to test the spatial model. The test results are presented in

Table 8. LM test of the spatial measurement model.

	Statistic	df	p-Value
Spatial error
Moran’s I	5.341	1	0.000
LM	25.980	1	0.000
Robust LM	1.939	1	0.000
Spatial lag			0.000
LM	25.612	1	0.000
Robust LM	1.571	1	0.000

and

Table 9. LR and Wald test results of model selection.

	Statistic	p-Value
LR-SDM-SAR	25.930	0.000
LR-SDM-SEM	26.760	0.000
Wald-sDM-SLM	46.23	0.000
Wald-SDM-SEM	37.26	0.000

.

It can be seen from the spatial test results in Table 8 and Table 9 that the LM test results suggest that the model can use a spatial econometric model while the LR test results indicate that the SDM model cannot degenerate into the SAR model and SEM model. Therefore, the relevant tests of the spatial econometric model confirm that this article is appropriate for the spatial Durbin model. The spatial Durbin econometric model is presented here:

$${CEI}_{it}={\mu }_0+\mathsf{\rho }{\mathrm{w}}_{\mathrm{ij}}{\mathrm{C}\mathrm{E}\mathrm{I}}_{\mathrm{it}}+\phi \ln {\mathrm{MS}}_{\mathrm{it}}+\phi w_{\mathrm{ij}}\ln {\mathrm{M}\mathrm{S}}_{\mathrm{it}}+\vartheta \sum _1^{p}X+\mathsf{\vartheta }{\mathrm{w}}_{\mathrm{ij}}\sum _1^{p}X+{\mathsf{\mu }}_{\mathrm{i}}{+\mathsf{\lambda }}_{\mathrm{t}}+{\epsilon }_{\mathrm{it}}$$

(10)

In this model, $\rho$, $\phi$, and $\vartheta$ denote the spatial regression coefficients, ${\mathrm{w}}_{\mathrm{ij}}$ denotes the spatial weight matrix, ${\mathsf{\mu }}_{\mathrm{i}}$ denotes the spatial effect, ${\mathsf{\lambda }}_{\mathrm{t}}$ denotes the time effect, and ${\mathsf{\epsilon }}_{\mathrm{it}}$ is the random disturbance term.

We finally choose the spatial Durbin model for empirical analysis. The benchmark results of the SDM model are shown in

Table 10. Direct and indirect effects of the spatial Durbin model.

Variables	Explained Variable: CEI
	Direct Effects	Indirect Effects	Total Effects
lnMS	0.358 ***	0.359 **	0.296
	(0.103)	(0.175)	(0.213)
lnHR	−2.320 *	−0.997	−3.317
	(1.242)	(2.764)	(3.147)
lnER	0.090 *	0.063	0.448 ***
	(0.050)	(0.087)	(0.118)
lnTI	0.147	−0.450	−0.303
	(0.183)	(0.463)	(0.509)
lnPD	5.961 ***	−3.701 ***	2.259 *
	(0.712)	(1.301)	(1.188)
Observations	540	540	540
Number of id	30	30	30
R-squared	0.128	0.128	0.128

Note. *, **, and *** indicate the levels of significance at 10%, 5%, and 1%, respectively, while () represents the standard error.

.

As can be seen from Table 10, market segmentation had significant spatial spillover effects on carbon emission intensity. This finding implies that market segmentation intensifies carbon emission intensity within the region and further strengthens the carbon emission intensity of neighboring regions. From the perspectives of total spatial effect, as well as direct and indirect spatial effects, market segmentation positively contributed to carbon emission intensity. However, the spatial total effect was not statistically significant and the spatial direct and indirect effects were highly significant. This study demonstrates that market segmentation affects carbon emission intensity within a specific region and the impact is extended to the neighboring areas, implying a strong presence of spatial spillover effects. This can be attributed to local governments’ pursuit of maximizing regional interests, which has intensified the fragmentation effect within a relatively closed and segmented market. This, in turn, has led to a proliferation of low-end industries yielding high returns and high emissions. This development pattern has increased energy consumption and resulted in substantial resource wastage, significantly amplifying carbon emission intensity.

Moreover, prominent obstacles among regions and market fragmentation and segmentation have hindered the formation of a cohesive and effectively regulated national market. Consequently, leveraging the scale effect of a large market has become difficult, thereby hindering the unrestricted flow of technology, labor, and innovation across regions. This has distorted resource allocation, reduced efficiency in energy utilization, and aggravated carbon emission intensity.

6. Conclusions and Policy Implications

6.1. Conclusions

Based on the theoretical analysis of the mechanism of market segmentation on carbon emission intensity, this paper employs panel data spanning from 2003 to 2020, encompassing 30 provinces in China. The aim is to empirically analyze the impact of market segmentation on carbon emission intensity through the utilization of the benchmark regression model, the intermediary effect model, and the spatial panel model. The econometric model constructed in this study aids in identifying viable development paths to achieve the “dual carbon goals”. The main findings of this study can be summarized as follows: firstly, market segmentation demonstrates a positive relationship with carbon emission intensity and these results remain robust, even after a series of tests to ensure validity and reliability; secondly, the heterogeneity analysis results indicate that market segmentation has a greater impact on carbon emission intensity in the eastern region than in the central and western regions. Over time, the promotion effect of market segmentation on carbon emission intensity gradually diminishes. Furthermore, the transmission mechanism reveals that resource misallocation and industrial structure are market factors that serve as mediating variables for the impact of segmentation on carbon emission intensity. Finally, the spatial effect analysis demonstrates that market segmentation exhibits a significant spatial correlation with carbon emission intensity. Market segmentation has a significant positive promotion effect on the carbon emission intensity of both local and neighboring areas, indicating a significant positive spatial spillover effect between market segmentation and carbon emission intensity.

6.2. Policy Suggestions

This study’s findings provide valuable insights for policymaking and practical guidance that can assist China in reaching its carbon peak and carbon neutrality goals. Drawing from the research findings mentioned above, this study aims to fulfill the “dual carbon goals” and presents beneficial policy implications.

First, market segmentation should be disrupted and regional synergy in governance should be promoted. Local governments should abandon local thinking, break down the barriers between regions, and eliminate market segmentation phenomena, including divisions and partitions. Furthermore, they should eliminate local protectionism, discard the development strategy and concept of “beggar-thy-neighbor,” and promote enhanced cross-regional exchanges and cooperation. This will enable the realization of complementary advantages and the establishment of a new, spatially linked, and efficient development paradigm. The central government should address market fragmentation caused by local governments and implement effective policy measures to enhance regulation and enforcement. Moreover, they should delegate appropriate authority to local governments to effectively combat market fragmentation while expediting the development of a comprehensive national unified market. The central and local governments should collaborate to reduce regional carbon emissions. They should clarify the roles and responsibilities of local governments in environmental protection and develop comprehensive policies for regional environmental governance. Strengthening these measures will enhance regional collaboration, thereby contributing to China’s low-carbon green development goals.

Second, industrial structure should be promoted. Local governments should work to eliminate regional barriers, facilitate coordinated industrial development, and phase out energy-intensive and polluting low-end industries. They should actively promote the transition of industries toward greener, more technologically advanced, and high-end sectors. Additionally, it is crucial to strengthen the mechanism for industrial transformation and upgrading and offer guidance and incentives to local governments to actively drive carbon emission reduction through industrial transformation and upgrading. For the efficient flow of resources and factors to sectors with high industrial production efficiency, leveraging regional development advantages and promoting coordinated industrial layouts between regions is essential. This approach allows us to achieve scale development in industries and harness positive external effects, including the “stem school,” knowledge spillover, technology spillover, and scale effects resulting from industrial transformation and upgrading, which can eventually accelerate the adoption of a new green development pattern with low carbon emissions in the region. Local governments should enhance incentives for low-carbon green development by utilizing supporting policies and funds to promote sustainable green development models, including low-carbon green technologies, clean energy, and other environmentally friendly approaches. Moreover, they should continuously improve low-carbon green applications through various methods to reduce the negative impact of carbon emissions.

Finally, it is essential to optimize and enhance the assessment criteria. Government involvement plays a crucial role in China’s pursuit of its “dual carbon goals,” with local governments consistently at the forefront. Thus, there is a need to continually refine the assessment criteria by placing greater emphasis on the green and low-carbon aspects of government performance evaluations. This involves increasing the weight of green criteria in assessments, shifting away from a sole focus on economic interests, and prioritizing sustainable and high-quality regional economic development while aligning low-carbon green policies with developmental requirements. Moreover, the government should explore and establish a cross-regional ecological and environmental compensation mechanism at a broader developmental level. This should include the introduction of compensation, incentives, and penalties tailored to local conditions for ecological and environmental management. The promotion and management mechanisms for local government officials should be continually optimized and improved to sustainably motivate the enthusiasm and proactiveness of local governments in pursuing green development. The ultimate goal is to effectively reduce regional carbon emissions, enhance regional environmental quality, and lay a solid foundation for achieving the “dual carbon targets.”

6.3. Limitations and Future Work

Nonetheless, while this study offers valuable insights, it is important to recognize certain limitations. Firstly, this study did not account for the impact of the COVID-19 epidemic. Subsequent research should comprehensively consider the effects of the COVID-19 epidemic on data changes to enhance the reliability of the analysis. Secondly, this study focused on several factors influencing carbon emission intensity; however, additional factors, like fiscal decentralization and fiscal vertical imbalance, might not have been considered. Therefore, addressing these issues in the future can provide valuable directions for further research.