A Study on the Decoupling Effect and Driving Factors of Industrial Carbon Emissions in the Beibu Gulf City Cluster of China

Abstract

This study investigates the decoupling relationship between industrial carbon emissions and economic development in the Beibu Gulf City Cluster based on panel data from 2005 to 2022. It also uses the Tapio decoupling model to assess the degree of decoupling and synergy in Guangdong, Guangxi, and Hainan and combines it with the logarithmic mean differential index (LMDI) decomposition model to study the driving factors affecting industrial carbon emissions. The study indicates that the industrial carbon emissions of the Beibu Gulf City Cluster increases from 71.42 MT in 2005 to 108.29 MT in 2022 but peaks in 2020 and changes from weak decoupling to strong decoupling; the synergistic relationship among Guangdong, Guangxi, and Hainan will evolve from poor to favorable. The LMDI decomposition results show that the economic scale and population scale effects increase 157.05 MT and 11.67 MT of carbon emissions in the study period, while the optimization of energy structure and energy intensity reduces 117.26 MT and 19.60 MT of carbon emissions, respectively, and the industrial development of many cities in the Beibu Gulf region gradually decouples economic growth and carbon emissions after 2021. Based on this, this study proposes targeted measures to reduce carbon emissions from industrial production in the Beibu Gulf City Cluster, which is of constructive significance for promoting sustainable industrial development in the region.

1. Introduction

With the ongoing process of industrialization, there has been a notable rise in CO₂ emissions [1]. The International Energy Agency (IEA) has identified fossil fuel combustion and industrial activities as the primary sources of CO₂ emissions, noting that fossil energy-driven economic growth has adversely affected atmospheric carbon accumulation [2]. As one of the world’s largest carbon emitters, China faces enormous challenges in reducing industrial carbon emissions while maintaining economic growth [3]. Energy conservation and carbon reduction is an important initiative to actively and steadily push forward carbon peaking and carbon neutrality goals and to promote a comprehensive green transformation of economic and social development [4]. Carbon reduction targets and measures have been incorporated into national- and local-level “dual-carbon” strategic plans to achieve this goal. In the case of the Beibu Gulf City Cluster, for example, the 14th Five-Year Plan for the Construction of the Beibu Gulf City Cluster, approved by the National Development and Reform Commission in 2022, proposes to “promote green and low-carbon development”, calling for the optimization of the energy structure, the development of a recycling economy, and the strengthening of ecological protection and restoration.

The Beibu Gulf City Cluster spans China’s Guangxi, Guangdong, and Hainan provinces and regions [5]. In recent years, due to China’s promotion of the construction of the Beibu Gulf City Cluster, the region has achieved certain advancements in industrialization; however, it remains in a developmental phase, confronting challenges such as a predominance of heavy industries, a deficiency of technology-intensive sectors, and a significant share of low-end products. Significant carbon emissions from industrial production processes inflict considerable harm on the ecological environment and adversely impact the industrial competitiveness and economic growth rate of urban areas. Conflicts between emission reduction objectives and economic growth are inevitable. Addressing the challenge of controlling and diminishing total industrial carbon emissions while ensuring sustained economic growth, as well as achieving alignment between carbon emission targets and economic performance, and strengthening industrial energy conservation and emission reduction in the Beibu Gulf City Cluster, constitutes a significant topic of inquiry [6].

Based on the above background, this paper examines the drivers of major industrial carbon emissions and the decoupling of industrial carbon emissions from regional economic development in the Beibu Gulf City Cluster while proposing viable low-carbon development strategies tailored to varying degrees of economic advancement and industrial carbon emissions to promote the process of sustainable industrial development in the Beibu Gulf City Cluster.

2. Literature Review

The main sources of carbon emissions consist of the burning of fossil fuels, industrial production, and agricultural activities [7]. Industrial processes, like cement, steel, and chemical manufacturing, produce substantial carbon emissions. To enhance the comprehension of the effects of carbon emissions, scholars have created many models for quantifying these emissions, including Integrated Assessment Models (IAMs), Climate Economic Models (CEMs), and Earth System Models (ESMs) [8]. Two prevalent methodologies employed to evaluate the determinants of carbon emissions are the Structural Decomposition Analysis (SDA) and the Index Decomposition Analysis (IDA) [9]; while the SDA requires more stringent data input, the IDA approach demonstrates advantages in operational flexibility due to its less data-intensive nature, with the Logarithmic Mean Divisia Index (LMDI) mode being a representative methodology in this category. The decoupling relationship between economic growth and carbon emissions has emerged as a critical research focus in sustainability science [10]. Originating from the conceptual framework developed by the Organization for Economic Cooperation and Development (OECD), this analytical paradigm underwent substantial methodological refinement through Tapio’s elasticity coefficient model [11], subsequently evolving into a widely adopted analytical tool.

In conjunction with the LMDI, scholars have dismantled the segmented drivers based on macro-decoupling. On the one hand, some scholars consider economic growth [12], increased energy consumption [13], and population explosion [14] as the most important factors affecting carbon emissions. They believe that in the early stages of industrialization, the rapid increase in energy demand will make carbon emissions grow faster than GDP growth, making it difficult to achieve green development. On the other hand, some scholars emphasize that energy efficiency improvements and energy mix shifts will reduce emissions and increase GDP growth. However, this approach is limited to the linear estimation of economic and carbon emission indicators, ignoring the nonlinear effects of interregional industrial divisions of labor, spatial spillovers, and the process of technology diffusion. It has been shown that positive feedback mechanisms for synergistic emission reduction and information sharing are more likely to be formed in regions with similar industrial structures and resource endowments [15], while synergistic effects are poorer among regions with large differences in resource structures [16]. Some regions tend to have problems of duplicated construction and excessive competition from energy-intensive industries due to a high degree of homogeneity in the industrial structure; others have large gaps in the depth and breadth of the application of clean energy technologies due to different infrastructures, financial environments [17], and talent pools.

Despite the enrichment of academic research on the decomposition and decoupling effects of carbon emission factors, the majority of studies predominantly concentrate on the decoupling relationship between carbon emissions and economic growth in a single region. Due to the inter-provincial nature of the Beibu Gulf City Cluster, i.e., there are great differences in the industrial structure, resource endowment, and development policies in Guangxi, Guangdong, and Hainan, ignoring the synergistic state of decoupling across regions makes it impossible to formulate scientific and reasonable carbon emission reduction policies and economic development strategies. Neglecting cross-regional decoupling synergies makes it impossible to formulate scientific and reasonable carbon emission reduction policies and economic development strategies for the region. To address this critical research gap, this study focuses on the Beibu Gulf City Cluster, a typical region across provinces and regions, and utilizes the decoupling elasticity coefficient and the cross-decoupling index to analyze the decoupling status and degree of synergy between economic growth and industrial carbon emissions in the Beibu Gulf City Cluster. By combining annual emission trends with factor decomposition, it proposes differentiated emission reduction paths to provide directional guidance for enhancing ecological protection and economic development in the Beibu Gulf City Cluster.

3. Methodology and Data Sources

3.1. Overview of the Research Area

The Beibu Gulf City Cluster is located on the southwest coast of China, covering several cities in Guangxi, Guangdong, and Hainan provinces (regions), and is an important gateway connecting China and ASEAN (Figure 1). There are significant differences in the development of the cluster: the Guangxi section is facing the carbon lock-in effect of heavy industries, the Guangdong section is in the stage of low-carbon transformation of traditional industries, and the Hainan section is exploring the new energy-first mode based on the policy advantages. In 2022, for example, industrial carbon emissions account for more than 65% of total carbon emissions in the Guangxi section, 30% in the Guangdong section, and less than 5% in the Hainan section. According to the 14th Five-Year Plan for the Construction of the Beibu Gulf City Cluster, the region is promoting structural reform through the “two horizontal and three vertical” spatial layout. The three core cities of Nanning, Zhanjiang, and Haikou are, respectively, responsible for regional financial centers, advanced manufacturing bases, and the opening hub of the free trade port, while the coastal economic corridor connects ports, such as Qinzhou and Fangchenggang, focusing on the development of green port development of green port.

3.2. Research Methodology

3.2.1. Measurement of Industrial Carbon Emissions in the Beibu Gulf City Cluster

Currently, the energy consumption method is commonly used to measure carbon emissions, and this study measured industrial carbon emissions in the Beibu Gulf City Cluster utilizing the carbon emission inventory estimating methodology established by the United Nations Intergovernmental Panel on Climate Change (IPCC); the formula is as follows [18]:

$$C=\sum E_i\times {CV}_i\times {CCF}_i\times {COF}_i\times 44/12$$

(1)

C is the total industrial carbon emission; $E_i$ is the ith energy consumption; ${CV}_i$ is the average low-level heating value of the ith energy source; ${CCF}_i$ is the carbon content of the ith energy source; ${COF}_i$ is the carbon oxidation factor of the ith energy source; and 44/12 denotes the molecular weight ratio of carbon dioxide to carbon. The energy consumption for each energy source can be multiplied by the standard coal conversion factor to convert it to standard coal consumption, thereby simplifying Equation (1) to Equation (2).

$$C=\sum E_i\times e_i\times p_i\times 44/12$$

(2)

$E_i$ is the consumption of energy source i; $e_i$ is the standard coal conversion factor of energy source i; and $p_i$ is the carbon emission factor of energy source i [19]. The parameters for calculating the carbon emissions of the nine energy sources are shown in

Table 1. Parameters for the calculation of carbon emissions.

Parameters	Raw Coal	Coke	Crude Oil	Petrol	Diesel	Diesel Fuel	Bunker Fuels	Electricity	Natural Gas
Standard coal conversion factor	0.7143	0.9714	1.4286	1.4714	1.4717	1.4571	1.4286	0.1229	13.330
Carbon emission factor	0.76	0.86	0.59	0.55	0.57	0.59	0.62	0.68	0.45

[20].

3.2.2. Tapio Decoupling Model

The Tapio decoupling model [11] was developed based on the OECD decoupling model, which overcomes the dilemma of the OECD decoupling model in the choice of the base period and improves the accuracy of the decoupling measure. This paper employed the Tapio decoupling model to compute decoupling elasticity, facilitating an analysis of the decoupling status and the degree of coordination between carbon emissions and economic development. The decoupling model of industrial carbon emissions and economic development for the Beibu Gulf City Cluster is represented in Equation (3):

$$T={\displaystyle \frac{∆C/C_0}{∆G/G_0}}$$

(3)

T is the decoupling index, $∆C$ is the change in carbon emissions, $∆G$ is the change in economic value added, and $C_0$ is the carbon emissions in the base period. Based on the values of ∆C, ∆G, and T, Tapio classified the decoupling states into three categories, weak decoupling, strong decoupling, and recessionary decoupling, and the corresponding numerical characteristics of the decoupling states are shown in Figure 2. The classification was subdivided into eight states. The specific decoupling criteria are listed in

Table 2. Standards for decoupling carbon emissions to economic growth.

Sorting of Decoupled States	Economic Growth Rate	Carbon Emission Growth Rate	Decoupling Elasticity Coefficient	Decoupled State
1	$∆G>0$	$∆C<0$	$T<0$	Strong decoupling
2	$∆G>0$	$∆C>0$	$0<T<0.8$	Weak decoupling
3	$∆G>0$	$∆C>0$	$0.8<T<1.2$	Expansive coupling
4	$∆G>0$	$∆C>0$	$T>1.2$	Expansive negative decoupling
5	$∆G<0$	$∆C>0$	$T<0$	Strong negative decoupling
6	$∆G<0$	$∆C<0$	$0<T<0.8$	Weak negative decoupling
7	$∆G<0$	$∆C<0$	$0.8<T<1.2$	Recessive coupling
8	$∆G<0$	$∆C<0$	$T>1.2$	Recessive decoupling

.

3.2.3. Measurement of the Cross-Decoupling Index

In order to be able to deeply analyze the synergistic relationship between economic development and resource environment among regions, this paper constructed a cross-decoupling index to evaluate the decoupling synergistic relationship between inter-regional economic development and resource consumption and waste emission:

$${\epsilon }_{rs}={\displaystyle \frac{T_{ir}}{T_{is}}}$$

(4)

where ${\epsilon }_{rs}$ denotes the cross-decoupling index between region r and region s; $T_{ir}$ and $T_{is}$ denote the decoupling elasticity coefficients of region r and region s in year i, respectively. This paper analyzed the decoupling synergistic relationship between economic growth and industrial carbon emissions among regions by determining the cross-decoupling index between regions. In order to better judge the degree of decoupling synergy, the state of decoupling synergy was divided into seven symmetric intervals according to ${\epsilon }_{rs}$ (

Table 3. Synergistic state of economic growth and industrial carbon emissions.

${\mathit{\epsilon }}_{\mathit{r}\mathit{s}}$	$(-\mathbf{\infty }$, 0.4)	[0.4, 0.6)	[0.6, 0.8)	[0.8, 1.25)	[1.25, 2.5)	[2.5, 5)	$[5,+\mathbf{\infty }$)
Cooperative State	Uncoordinated	General Coordination	Coordinated	High Degree of Coordination	Coordinated	General Coordination	Uncoordinated

).

3.2.4. Decomposition of Industrial Carbon Emission Drivers

In 1989, at the IPCC seminar, Prof. Yoichi Kaya [21] proposed the Kaya equation to decompose the factors influencing carbon emissions.

$$C={\displaystyle \frac{C}{E}}\times {\displaystyle \frac{E}{G}}\times {\displaystyle \frac{G}{P}}\times P$$

(5)

C denotes total industrial carbon emissions, E denotes total energy consumption, G denotes GDP, and P denotes population.

This paper combined the actual situation of industrial production activities; the kaya constant equation was appropriately deformed and quantitatively decomposed concerning the Logarithmic Mean Divisia Index (LMDI) model [22]. The LMDI method is widely used in the study of carbon emission drivers, and its advantage lies in the fact that there is no residual value term after the factor decomposition, which can decompose the factor completely. Based on Kaya’s constant equation, taking into account the actual situation of the study area and data availability, energy structure, energy consumption intensity, economic scale, and population scale were selected for the study, and the decomposition equations presented in this study are as follows:

$$C=\alpha \times \beta \times \gamma \times p$$

(6)

$\alpha$ = C/E represents the emission intensity, indicating the characteristics of the energy structures; $\beta$ = E/G signifies the GDP energy intensity, which illustrates the economy’s energy efficiency and production structure at a macroeconomic level; and $\gamma$ = G/P represents GDP per capita, indicating the impact of economic size, where P is the population. The factors influencing carbon emissions are represented by $\alpha$, $\beta$, $\gamma$, and $p$, which correspond to the energy consumption structure, energy intensity, economic scale, and population scale, respectively.

This study employed the LMDI model to decompose the factors influencing carbon emissions in the Beibu Gulf City Cluster [23]. The aggregate contribution of industrial carbon emissions is $∆C$, The impact of energy structures, energy intensity, economic scale, and population scale on carbon emissions is represented by ${∆C}_{\alpha },{∆C}_{\beta }{,∆C}_{\gamma },{ ∆C}_p$. The influence of each component on the carbon emissions of the Beibu Gulf City Cluster can be expressed as follows:

$$∆C={∆C}_{\alpha }+{∆C}_{\beta }{+∆C}_{\gamma }+{∆C}_p=C^t-C^0$$

(7)

$${∆C}_{\alpha }=\sum {\displaystyle \frac{C^t-C^0}{{lnC}^t-{lnC}^0}}ln{\displaystyle \frac{{\alpha }^t}{{\alpha }^0}}$$

(8)

$${∆C}_{\beta }=\sum {\displaystyle \frac{C^t-C^0}{{lnC}^t-{lnC}^0}}ln{\displaystyle \frac{{\beta }^t}{{\beta }^0}}$$

(9)

$${∆C}_{\gamma }=\sum {\displaystyle \frac{C^t-C^0}{{lnC}^t-{lnC}^0}}ln{\displaystyle \frac{{\gamma }^t}{{\gamma }^0}}$$

(10)

$${∆C}_p=\sum {\displaystyle \frac{C^t-C^0}{{lnC}^t-{lnC}^0}}ln{\displaystyle \frac{p^t}{p^0}}$$

(11)

The value of $∆C={∆C}_{\alpha }+{∆C}_{\beta }{+∆C}_{\gamma }+{∆C}_p$ with $T={\displaystyle \frac{∆C/C_0}{∆G/G_0}}$ represents the association between the two equations, which enables the decoupling elasticity of industrial carbon emissions from GDP to be separated into decoupling elasticities of the four variables.

$$T={\displaystyle \frac{({∆C}_{\alpha }+{∆C}_{\beta }{+∆C}_{\gamma }+{∆C}_p)/C}{∆G/G}}={\displaystyle \frac{{∆C}_{\alpha }/C}{∆G/G}}+{\displaystyle \frac{{∆C}_{\beta }/C}{∆G/G}}+{\displaystyle \frac{{∆C}_{\gamma }/C}{∆G/G}}+{\displaystyle \frac{{∆C}_p/C}{∆G/G}}={\epsilon }_{\alpha }+{\epsilon }_{\beta }+{\epsilon }_{\gamma }+{\epsilon }_p$$

(12)

In this study, the energy structure decoupling elasticity (denoted by ${\epsilon }_{\alpha }$), energy intensity decoupling elasticity (denoted by ${\epsilon }_{\beta }$), economy size decoupling elasticity (denoted by ${\epsilon }_{\gamma }$), and population size decoupling elasticity (denoted by ${\epsilon }_p$) were used as indicators.

3.3. Data Sources and Processing

Due to the unavailability of statistical data for Yangjiang, Danzhou, and Dongfang, the remaining prefecture-level cities in the Beibu Gulf City Cluster were selected as the research focus. Data on energy consumption were primarily sourced from the Guangxi Statistical Yearbook, the Guangdong Statistical Yearbook, and the statistical yearbooks of prefectural cities; carbon emission factor data were derived from the China Energy Statistical Yearbook and the Urban Greenhouse Gas Inventory Study; and GDP and other socio-economic statistics for prefectural administrative units were obtained from provincial and municipal statistical yearbooks and bulletins on national economic and social development. Considering the influence of inflation and to ensure the comparability of the data, they were adjusted using comparable prices for the 2005 base period. Missing data were replaced with fitted values derived from linear interpolation, and the relevant data were preprocessed before computation to meet the accuracy of the calculations.

4. Results and Discussion

4.1. Trends in Industrial Carbon Emissions in the Beibu Gulf City Cluster

From 2005 to 2022, the GDP and industrial carbon emissions of the Beibu Gulf City Cluster exhibited divergent trends (Figure 3); although the total industrial carbon emissions still grow, the growth rate is lower than the regional economic growth rate, indicating that the effectiveness of the existing emission reduction measures has not yet been fully realized. The industrial carbon emissions of the Beibu Gulf City Cluster increased from 71.42 Mt in 2005 to 108.29 Mt in 2022, reflecting a relatively modest growth overall. Since 2011, carbon emissions have stabilized, and from 2015 to 2022, industrial carbon emissions in the Beibu Gulf City Cluster have remained largely constant, with some years experiencing a decline due to the local government’s proactive adherence to the national strategy for energy conservation and emission reduction. The total GDP has exhibited a consistent and significant growth trend, especially accelerating after 2015, while carbon emissions have stabilized at a low level, progressing towards the objective of a low-carbon transition. This trend indicates that the Beibu Gulf City Cluster has progressively attained efficient regulation of industrial carbon emissions while fostering a sustainable industrial development trajectory alongside swift economic expansion.

A specific analysis of industrial carbon emission data for the Beibu Gulf City Cluster over 17 years (2005–2022) shows that industrial carbon emissions surged in the early years of accelerated industrialization and then declined after stabilization (Figure 4). In particular, carbon emissions peaked between 2014 and 2017, when industrialization accelerated, before stabilizing or even declining slightly. Cities like Zhanjiang, Maoming, and Nanning have seen swift industrialization, marked by a surge in industrial businesses, enlarged production scales, and substantial increases in carbon emissions. The development of high-energy-consuming enterprises has been rigorously regulated by the governments of Zhanjiang and Maoming since 2017, and the pace of industrial restructuring has been accelerated. Consequently, the overall growth rate of industrial carbon emissions has slowed. Beihai, Fangchenggang, and Qinzhou are situated on the shores of the Beibu Gulf and possess abundant wind resources, which provide a large amount of clean energy for industrial enterprises, resulting in relatively low carbon emissions that are growing slowly.

4.2. Analysis of the Decoupling of Industrial Carbon Emissions from Economic Development

4.2.1. Decoupling State Analysis

The Tapio decoupling model can be analyzed to show that the period 2005–2014 was a transition period for the low-carbon transition of the industry in the Beibu Gulf City Cluster, and the Beibu Gulf City Cluster first showed a strong decoupling state in 2007, indicating that the economic development of the region began to gradually remove its dependence on high carbon emissions. After 2014, with the implementation and continuous promotion of offshore wind power and other renewable energy projects as well as the policy of the Beibu Gulf Port’s ‘14th Five-Year Plan’ for Green Port Development, the dependence of industrial production on traditional high-carbon energy sources has been significantly suppressed, with strong decoupling occurring in more than half of the years. Especially after 2020, there are three consecutive years of strong decoupling, and according to the elasticity coefficient of the Tapio model, the growth rate of industrial carbon emissions has slowed down significantly during this period and even showed a downward trend, which further proves that the Beibu Gulf City Cluster has made substantial progress in green development and industrial upgrading (Figure 5).

4.2.2. Vertical Perspective

In this study, cities with ten or more years of strong decoupling were classified as advantageous zones, whereas those with eight years or less were categorized as underperforming zones. In between is the better zone.

Table 4. Types of industrial carbon decoupling in the Beibu Gulf city cluster, 2005–2022.

City	Strong Decoupling	Weak Decoupling	Expansive Coupling	Strong Coupling	Strong Negative Decoupling
Zhanjiang	10	5	-	3	-
Maoming	10	4	2	1	1
Nanning	9	8	1	-	-
Beihai	8	7	1	1	1
Fangchenggang	9	7	2	-	-
Qinzhou	9	8	-	1	-
Yulin	9	6	1	2	-
Chongzuo	8	8	2	-	-
Haikou	9	7	2	-	-

indicates that Zhanjiang and Maoming are the advantageous zones for the decoupling effect of industrial carbon emissions. Zhanjiang has been in strong or weak decoupling for 15 years between 2005 and 2022 and has not experienced strong negative decoupling. Maoming, on the other hand, has been in strong or weak decoupling for a cumulative total of 14 years and has made the transition from strong negative decoupling to an expansive coupling state to strong decoupling, and the type of strong decoupling has been stable for five consecutive years. Nanning, Fangchenggang, Qinzhou, Yulin, and Haikou are in a better state of industrial carbon decoupling, with strong and weak decoupling over 15 years, but still have some risk of negative decoupling. For example, Yulin and Qinzhou had two and one instances of negative decoupling states, respectively, suggesting that they need to further balance the intensity of industrial carbon emissions and the rate of economic growth.

4.2.3. Synergistic Analysis

Table 5. The cross-decoupling index and synergistic state of economic growth and industrial carbon emissions in the Beibu Gulf City Cluster, 2005–2022.

Year	${\mathit{\epsilon }}_{12}$	${\mathit{\epsilon }}_{13}$	${\mathit{\epsilon }}_{23}$	Guangdong/Guangxi	Guangdong/Hainan	Guangxi/Hainan
2005	3.143	6.036	1.921	general coordination	uncoordinated	coordinated
2006	−3.467	−0.201	0.058	uncoordinated	uncoordinated	uncoordinated
2007	−6.333	4.425	−0.699	uncoordinated	general coordination	uncoordinated
2008	5.386	−4.467	−2.686	uncoordinated	uncoordinated	uncoordinated
2009	−4.216	3.058	−0.063	uncoordinated	general coordination	uncoordinated
2010	4.063	1.932	0.476	general coordination	coordinated	general coordination
2011	−0.094	−0.098	1.051	uncoordinated	uncoordinated	high degree of coordination
2012	−0.966	−1.018	1.054	uncoordinated	uncoordinated	high degree of coordination
2013	−4.229	−6.368	0.147	uncoordinated	uncoordinated	uncoordinated
2014	9.209	1.776	0.193	uncoordinated	coordinated	uncoordinated
2015	3.419	1.533	0.448	general coordination	coordinated	general coordination
2016	0.295	0.389	1.317	uncoordinated	uncoordinated	high degree of coordination
2017	0.038	−0.005	−0.140	uncoordinated	uncoordinated	uncoordinated
2018	−9.163	1.082	−0.118	uncoordinated	high degree of coordination	uncoordinated
2019	−9.548	−6.000	0.628	uncoordinated	uncoordinated	coordinated
2020	1.832	0.956	0.522	coordinated	high degree of coordination	general coordination
2021	1.501	0.954	0.636	coordinated	high degree of coordination	coordinated
2022	2.980	4.854	1.630	general coordination	general coordination	coordinated

illustrates that from 2005 to 2019, the economic growth and industrial carbon emissions of Guangdong and Guangxi prefecture-level cities within the Beibu Gulf City Cluster predominantly exhibited uncoordinated characteristics. However, post-2020, these metrics gradually converge towards a coordinated state, suggesting an optimization of the industrial and economic structures in both regions and an initial success in the energy transition. From 2009 to 2013, as the economies of Guangdong and Hainan progressed and increasingly depended on the environment, their decoupling synergies enhanced yet remained unstable, with a cross-section of uncoordinated, coordinated, and general coordination states in some years. Between 2005 and 2009, the synergistic relationship between Guangxi and Hainan was poor due to disparities in industrial development speed. However, from 2009 onwards, the synergistic state gradually improved, and there were only a few years in which there was the uncoordinated state. This indicates that these two regions have integrated environmental protection into their economic development, enhancing their industrial competitiveness, thus serving as a model and leader for other regions and fostering the coordinated development of the regional economy and environment. The overall trend shows that the provinces (regions) within the Beibu Gulf City Cluster have experienced a shift in economic–environmental synergy from no synergy to general or even high synergy. By quantifying the specific data on resource inputs and environmental performance, it can be seen that the intensity of carbon emissions and energy consumption have both been on a downward trend in recent years, and the negative impacts of economic growth on the environment have gradually weakened.

4.3. Driver Analysis

The equations presented (Equations (5)–(12)) synthesize total industrial carbon emissions, aggregate energy consumption, gross regional product, and the population of the Beibu Gulf City Cluster to examine the four driving factors and their associated decoupling elasticities: energy structures, the energy intensity, the economic scale, and the population scale. The data are presented in

Table 6. The decomposition of factors affecting industrial carbon emissions in the Beibu Gulf City Cluster, 2005–2022.

Foundation Period	End Period	Total Effect	Energy Structure	Energy Intensity	Size of Economy	Size of Population
2005	2006	4.030	−6.474	0.018	9.093	1.403
2006	2007	0.682	−10.157	−2.163	12.303	0.699
2007	2008	−0.960	−10.371	−2.522	11.258	0.675
2008	2009	7.745	1.137	−0.705	6.840	0.473
2009	2010	4.607	−3.612	−7.218	17.902	−2.462
2010	2011	3.706	−10.548	−2.530	15.895	0.888
2011	2012	−2.274	−14.328	4.449	6.835	0.769
2012	2013	5.995	−1.411	−1.855	8.502	0.759
2013	2014	3.420	−0.328	−4.432	7.464	0.716
2014	2015	−5.263	−2.795	−8.477	5.205	0.804
2015	2016	6.532	−2.326	0.918	7.046	0.895
2016	2017	8.118	−0.043	−1.802	8.860	1.101
2017	2018	−4.482	−4.679	−11.092	10.291	0.999
2018	2019	6.567	2.710	−0.758	3.698	0.917
2019	2020	0.049	−3.298	0.092	1.063	2.191
2020	2021	−2.795	−22.748	2.915	16.517	0.520
2021	2022	−2.208	−14.467	6.870	5.181	0.208
2022	2023	−1.604	−13.522	8.699	3.105	0.115
Cumulative Contribution		31.865	−117.259	−19.599	157.053	11.671

and

Table 7. The decoupling elasticity of industrial carbon emission drivers in the Beibu Gulf City Cluster.

Foundation Period	End Period	${\mathit{\epsilon }}_{\mathit{\alpha }}$	${\mathit{\epsilon }}_{\mathit{\beta }}$	${\mathit{\epsilon }}_{\mathit{\gamma }}$	${\mathit{\epsilon }}_{\mathit{p}}$
2005	2006	0.044	−0.731	0.826	0.141
2006	2007	−0.058	−0.662	0.891	0.027
2007	2008	−0.005	−1.172	0.869	0.020
2008	2009	0.176	−0.045	0.964	0.071
2009	2010	0.017	−0.343	0.928	0.053
2010	2011	−0.082	−0.373	0.893	−0.002
2011	2012	−0.215	1.142	0.982	−0.010
2012	2013	−0.194	−0.266	0.912	−0.002
2013	2014	0.599	−2.231	1.135	−0.005
2014	2015	−0.380	−1.201	0.904	−0.010
2015	2016	−0.962	7.977	1.021	−0.006
2016	2017	−0.160	0.749	0.979	−0.003
2017	2018	−0.096	−0.813	0.944	−0.001
2018	2019	0.590	−1.445	1.062	−0.003
2019	2020	−3.115	12.086	0.923	−0.008
2020	2021	−0.245	1.0122	0.873	0.047
2021	2022	−0.384	1.697	0.952	0.015
2022	2023	−0.723	4.060	0.701	0.276

, respectively.

Economic and population size are the two major positive drivers of industrial carbon emissions in the Beibu Gulf City Cluster. During the study period, the economic and population size cumulatively increased about 157.053 megatons and 11.67 megatons of CO₂ emissions, respectively, indicating that the negative externalities of the expansion of regional industrial production and urbanization on the environment are still significant. As the most important factor driving the growth of carbon emissions, economic size increased CO₂ emissions by 15.895 Mt and 8.86 Mt in 2010 and 2017, mainly due to the dependence of the existing industrial model on fossil energy, and the increase in the scale of industrial production will inevitably result in a rise in carbon emissions; it is foreseeable that this factor will continue to be the most significant factor positively driving carbon emissions in the future. According to Table 7, the elasticity of economic size to carbon emissions is mostly positive and above 0.8, indicating an “expansive coupling” in most years.

The impact of population size on carbon emissions is relatively small but also generally positive. In 2019, population size led to an increase in carbon emissions of 2.191 MT, suggesting that population growth expands consumption demand and production activities, which indirectly push up industrial carbon emissions. The decoupling elasticity of population size in Table 7 is lower than that of economic size during the study period, but it is also mostly in the fluctuating stage between expansive coupling or strong decoupling, indicating that it is still necessary to alleviate the pressure of emissions caused by population growth in the future by strengthening urban planning and energy-saving promotion.

Energy structures and intensity adversely affected the increase in carbon emissions during the study period, leading to cumulative reductions of 117.259 Mt and 19.60 Mt of carbon emissions, respectively. The energy mix decoupling elasticity is predominantly negative, indicating that the adjustment of energy structures plays a significant role in suppressing carbon emissions in most years. The change in energy structures has led to a decrease in carbon emissions of 14.33 Mt in 2011 and 22.75 Mt in 2020. This suggests that restructuring the energy sector diminished reliance on fossil fuels for industrial production. Except for a few years in which carbon emissions increased, energy intensity predominantly significantly negatively impacted carbon emissions in most years, especially in 2007 and 2013, with reductions of 2.52 Mt and 4.43 Mt, respectively. This signifies that technical progress has diminished the energy consumption per unit of GDP, thereby restraining the increase in carbon emissions. These two factors are likely to remain central to future industrial carbon-reduction efforts. The overall trend shows that technological progress and the optimization of energy structure are the key paths for industrial carbon reduction in the region, and these two factors will remain the main directions for industrial carbon reduction efforts in the future.

For the four drivers of industrial carbon emissions in the Beibu Gulf City Cluster, the segmentation analysis of carbon emission drivers during the study period was conducted based on the LMDI decomposition model, and the results are shown in Figure 6. Regarding (1) 2005–2010, the initial development stage, economic and population size effects play a major role in promoting industrial carbon emissions. Industrial enterprises in the Beibu Gulf City Cluster are in the stage of rough development, and the expansion of the size of the employed population leads to the rapid growth of industrial carbon emissions. The energy structure effect alternates between positive and negative, mainly because coal and oil still dominate, and the proportion of clean energy is low, resulting in an unstable energy structure effect. Regarding (2) 2011–2016, the economic and population expansion stage, along with the expansion of the economy and the growth of population and demand, industrial energy demand has shown greater rigidity, pushing industrial carbon emissions up continuously. If mandatory emission reduction policies are enforced at this time, they will inevitably come at the expense of industrial economic development. Regarding (3) 2017–2022, the low-carbon transformation stage, as the Beibu Gulf City Cluster strengthens the emission reduction and transformation of key industrial sectors, the suppression of industrial carbon emissions by energy structure effects is gradually increasing. For example, the Beibu Gulf City Cluster Development Plan, which will be jointly revised by the National Development and Reform Commission and the Ministry of Housing and Construction in 2022, provides policy support for the green transformation of industry and the development of a low-carbon and recycling economy, which will help to further promote the low-carbon transformation of industrial enterprises and strengthen the joint prevention and control of the ecological environment.

5. Conclusions and Recommendations

5.1. Conclusions

By measuring the industrial carbon emissions data of the Beibu Gulf City Cluster from 2005 to 2022, the study verifies the decoupling relationship and synergy between carbon emissions and economic development while also examining the driving factors of industrial carbon emissions within the region. The following conclusions are derived:

From 2005 to 2022, the Beibu Gulf City Cluster has exhibited an upward trend in both industrial carbon emissions and economic development. Overall, the decoupling state of the Beibu Gulf City Cluster shows a transition from weak to strong decoupling, indicating that economic expansion is positively correlated with carbon emissions, but emission reduction factors are already at work. After 2020, the region entered a state of strong decoupling, and the growth rate of industrial carbon emissions slowed down or even decreased while the economy continues to grow, indicating that the low-carbon transformation of industries in the Beibu Gulf urban agglomeration is gradually showing results and that the increase in industrial carbon emissions has been under control, and the results of this study confirm the feasibility and necessity of the transformation of industrial production from high carbon to low carbon.

The overall synergistic state of economic growth and industrial carbon emissions between Guangdong and Guangxi, Guangdong, and Hainan in the Beibu Gulf City Cluster was poor, suggesting that they have not formed a good cooperation mechanism in the research and development of energy-efficient and emission-reducing technologies as well as in the advancement and utilization of clean energy, whereas the synergistic state between Guangxi and Hainan has stabilized since 2018, reflecting the role of industrial structure and factor endowment in the promotion of trans-regional emission reduction and economic integration, which corresponds to the viewpoints of scholars that “regional industrial synergy and environmental governance require deep complementarity at the factor level”. To further realize cross-regional synergistic decoupling, it is necessary to break down inter-provincial resource and technology barriers and form a pattern of cleaner production and market interconnection.

The decomposition of industrial carbon emission drivers through the LMDI shows that with the introduction of clean energy and the improvement of energy efficiency, the energy structure and energy intensity of the Beibu Gulf City Cluster has a more and more significant inhibitory effect on industrial carbon emissions. The expansion of the economic and population scale has a strong effect on industrial carbon emissions in the short term, and the structural problems of high-energy-consuming and high-emission industries still need to be solved in-depth, as heavy industries still dominate in the Beibu Gulf City Cluster; the agglomeration effect of capital-intensive and technology-intensive industries has not yet been formed.

However, this study still has the following shortcomings: First, this study focuses on industrial CO₂ emissions and does not comprehensively consider other air pollutants, water pollution, and solid waste emissions, thus failing to reflect the comprehensive effect of multi-pollutant synergistic management on economic development. Second, the Tapio-LMDI model used in this study has some limitations. The static perspective of the Tapio model makes it difficult to capture the dynamic characteristics of regional industrial transformation, and the LMDI decomposition results are subject to the subjective influence of the choice of driving factors, which may underestimate the hidden variables, such as technological progress. To address the above shortcomings, future scholars may consider incorporating multiple pollutant emissions or ecological footprint indicators into the framework of a decoupling analysis and adopting a multi-objective or multi-pollutant integrated management perspective to assess the comprehensive impact of the industrial sector on the environment. Econometric methods such as the Generalized Method of Moments (GMM) and the Difference–Differences (DID) method can be introduced to better control the model’s endogeneity and comprehensively assess the impacts of industrial transformation, carbon tax policy, and technological progress on the evolution of decoupling.

5.2. Recommendations

In light of the aforementioned conclusions and the decoupling status of the Beibu Gulf City Cluster, along with a differentiated study of the driving variables, specific recommendations for reducing carbon emissions from industrial output in the Beibu Gulf City Cluster are presented:

Regarding the green transformation of heavy industry in the harbor, combining the Tapio decoupling model with the LMDI decomposition results shows that the decoupling state of industrial carbon emissions and economic growth depends largely on the optimization of energy intensity and structure. The Beibu Gulf City Cluster hosts numerous port-dependent heavy industries. It is essential to advance intelligent transformation with digital technologies in these businesses. Taking the iron and steel industry as an example, the application of big data and artificial intelligence technology to precisely control energy inputs reduces energy wastage due to instability in the production process and promotes economic growth while further consolidating the stability of the decoupled state.

Regarding energy auditing and benchmarking, mandatory energy audits have been carried out for industrial enterprises in the Beibu Gulf urban agglomeration, and an energy management information platform has been established to provide real-time access to enterprise energy consumption data, facilitating energy monitoring and analysis by government regulators and corporations. However, attention needs to be paid to the current differences in digitization infrastructure among enterprises, and more socio-economic variables (e.g., enterprise size, nature of ownership) could be included to compensate for the lack of reliance on a single energy consumption statistical indicator.

Regarding offshore wind development and utilization, according to the decomposition of the drivers of industrial carbon emissions in the Beibu Gulf City Cluster, the optimization of energy structure has a potential long-term inhibiting effect in reducing industrial carbon emissions. The Beibu Gulf possesses abundant wind energy resources, enabling the government to entice robust firms to participate in developing large-scale offshore wind power projects via concession bidding and other strategies. Simultaneously, developing and building an offshore wind power equipment manufacturing industrial park will provide a comprehensive industry chain encompassing the manufacturing, installation, operation, and maintenance of wind power equipment, thereby supplying reliable, clean energy for industrial output.

Regarding retrofitting industrial parks for recycling, for example, in the Qinzhou Harbor Economic and Technological Development Zone, a water reuse system has been established within the zone to reduce freshwater withdrawals and sewage discharges. Leveraging synergies with neighboring cities or other industrial parks, it promotes the scaled operation of pollution control and resource recycling to improve overall efficiency.