Low-Carbon Transformation in Megacities: Benefits for Climate Change Mitigation and Socioeconomic Development—A Case Study of Shenzhen, China

Abstract

Megacities are the main sources of carbon emissions and are pillars of socioeconomic development due to economic prosperity, industrial development, and population agglomeration. Taking the megacity of Shenzhen, China, as an example, this research explored the advantages of low-carbon transformation in both climate change mitigation and socioeconomic progress. Soft linking of the Low Emissions Analysis Platform (LEAP) model with input–output analysis demonstrates that adopting feasible low-carbon transformation strategies has the potential to reverse the ongoing trend of carbon emission growth in Shenzhen, resulting in a peak before 2025. The peaks in carbon intensity and carbon emissions per capita occurred earlier than those in total carbon emissions. Furthermore, a total low-carbon investment of 462.04 billion CNY could yield approximately 799.49 billion CNY in output, 311.42 billion CNY in value-added, and 156.10 billion CNY in resident remuneration and create 1.79 million job opportunities during policy implementation. Taking into account both the reduction in carbon emissions and the socioeconomic benefits of low-carbon transformation, this research elucidates a potential pathway for Shenzhen to achieve synergy between mitigating climate change and promoting socioeconomic development, thus offering a valuable model for cities facing similar challenges.

1. Introduction

The sixth IPCC report [1] highlights the anthropogenic activities, particularly greenhouse gas emissions, that resulted in a global surface temperature increase of 1.1 °C above pre-industrial levels from 2011 to 2020. Climate change has led to widespread adverse impacts and irreversible losses to both nature and humanity [1]. In response, 195 countries reached a consensus at the Paris Climate Summit in 2015 to reduce greenhouse gas emissions. Furthermore, in September 2020, President Xi Jinping announced ambitious targets, known as the “dual-carbon” goals for China, aiming to achieve a carbon peak by 2030 and carbon neutrality by 2060. However, global greenhouse gas emissions have continued to increase, with energy-related CO₂ emissions reaching a record high of 37.4 Gt in 2023 [2]. The investment in the green economy is a main instrument to constrain carbon emissions and reach the climate neutrality target [3].

Cities play a significant role in mitigating climate change due to their substantial energy consumption and carbon emissions [4,5]. Energy-related carbon emissions in cities are projected to increase by 1.8% annually, causing the proportion of carbon emissions from cities to increase from 71% to 76% from 2006 to 2030 [6]. This trend can be largely attributed to the proliferation of industries and the growth of the population in cities [7].

Megacities, defined as those with populations of 10 million or more [8], have gradually become a major trend in urban development due to rapid urbanization. More than two-thirds of the world’s population is projected to be urbanized, with up to 90% of the increase centered in Asia and Africa [8]. The number of megacities increased from 6 in 1980 to 35 in 2020, with the average population increasing from 16.4 million to 25.8 million [9]. The United Nations Organization further highlights that the number of megacities may increase to 43 by 2030 [10].

In the context of global climate change and the rapid development of cities, megacities play a significant role not only in promoting socioeconomic development but also in mitigating climate change [11,12]. Existing research has provided abundant knowledge concerning the low-carbon transformation of megacities; most of it focuses on understanding the driving forces of carbon emissions and on making carbon emission projections. However, few studies have systematically explored the potential co-benefits of climate change mitigation and socio-economic progress resulting from a low-carbon transformation. To fill this research gap, this study assessed the sector-specific carbon reduction potential and the socioeconomic impacts of low-carbon investment based on a novel approach: a soft-linked model that integrates the LEAP model with an empirical IO model. This study aims to outline a pathway to reduce carbon emissions while maximizing socioeconomic benefits in megacities.

The marginal contributions of this study are expected to be threefold. First, we constructed the LEAP–Shenzhen model, which dynamically simulates carbon emissions in Shenzhen from 2020 to 2030 and analyzes the potential reduction in carbon emissions through effective low-carbon measures. Second, using input–output analysis (IOA), this study evaluated the socioeconomic benefits of low-carbon investment in Shenzhen, including increased output, value-added, residential income, and job opportunities. Third, by comparing the results of the LEAP and IO models, this study aims to identify the optimal pathway for Shenzhen to achieve synergy between carbon emission reduction and socioeconomic improvement, serving as a model for cities facing similar dilemmas, particularly those in developing countries.

The remainder of this paper is organized as follows: Section 2 offers a concise review of the literature on megacities, focusing on their characteristics and low-carbon transformation. Section 3 introduces the case study, methodology, and data, detailing the construction of the LEAP–Shenzhen and IO–Shenzhen models. Section 4 presents the results and a subsequent discussion and Section 5 presents the conclusions drawn from the study and explores policy implications.

2. Literature Review

With growing concerns regarding climate change and rapid urbanization, the low-carbon transformation of megacities has garnered considerable academic attention. This section reviews the literature on the characteristics and low-carbon transformation of megacities, aiming to identify potential research gaps concerning the cobenefits of climate change mitigation and socioeconomic progress resulting from low-carbon transformation in megacities.

2.1. Research on the Characteristics of Megacities

Megacities have become a major trend in urban development due to population growth, advancements in the finance, business, and transportation sectors, telecommunication networks [13], market forces, and job matching [9].

Megacities demonstrate unique development dynamics and a complex interaction of diverse demographic, social, political, economic, and ecological processes, setting them apart from large cities. Megacities are characterized by diversified industrial structures [13], higher income levels, higher GDP, and increased technological innovation and productivity [14]. However, rapid population growth in megacities often outpaces infrastructural development, resulting in fragmented and uncontrolled urban sprawl. This phenomenon contributes to increased traffic volumes and higher concentrations of industrial production, ecological strain, and environmental degradation [15]. In addition, megacities face greater risks of climate change due to their location, making them particularly susceptible to the effects of climate change. They also exacerbate existing vulnerabilities related to environmental pollution, capacity overload, and resource consumption [15].

2.2. Research on Low-Carbon Transformation in Megacities

Research on low-carbon transformations at the city level can be broadly classified into three categories. The first aims to identify the key factors driving carbon emissions in cities. The second stream focuses on developing comprehensive energy-economic models to quantify carbon emissions and simulate future dynamics. The third stream explores the wide-ranging socioeconomic impacts of low-carbon strategies. Together, these streams complement each other and provide a holistic understanding of the challenges and opportunities associated with low transformation in megacities.

2.2.1. Driving Forces behind Carbon Emissions

The driving forces behind carbon emissions in megacities vary depending on the development period and research methods employed [6,16]. The existing literature identifies the key factors that contribute to carbon emissions from three main aspects. The first is the consumption effect, such as market and energy consumption [17]. The second aspect is the growth effect, which includes growth in manufacturing, population size, land size, and GDP [7,18,19]. The third aspect is the development effect, which encompasses capital formation, higher living standards, and GDP per capita [18,20,21].

Three primary research methods are used extensively to investigate these factors. The first method is decomposition analysis, which includes techniques such as Kaya index decomposition [7] and the logarithmic mean Divisia index [18]. The second method encompasses economic approaches such as regression analysis [17,21] and the IO method [20]. The third method involves the innovative integration of computer science techniques, such as XG-Boost algorithms, into the research area [19].

2.2.2. Quantification and Scenario Analysis

Quantifying the future carbon emissions of megacities and exploring pathways toward carbon peaking and neutrality through scenario analyses have long been of academic interest. Hu [22] and Li [4] conducted scenario analyses to propose optimal paths for carbon peaking and neutrality in Shanghai using dynamic computable general equilibrium (DCGE) modeling and a LEAP model, respectively. Yang [23] simulated the emissions from six energy sectors in Ningbo, China. Huo [24] integrated the LEAP model into a system dynamics framework to investigate carbon emissions peaking in the residential building sector in Chongqing. Dong [25] introduced a life cycle assessment to quantify carbon emissions from urban public transport in Shenzhen and simulated three scenarios to evaluate future emissions and their reduction potentials. Hu [26] used the LEAP model to construct scenarios and policies for sustainable energy development in Shenzhen, China.

Four primary models are generally used to quantify and simulate future carbon emissions: top-down models, such as the Computable General Equilibrium (CGE) and IO models; bottom-up models, including the Market Allocation model (MARKAL), Model for Energy Supply Strategy Alternatives and their General Environmental Impact (MESSAGE), and LEAP; hybrid models, such as CGE-MARKAL; and Integrated Assessment Models (IAMs), such as the Dynamic Integrated model of Climate and the Economy (DICE) and Regional Integrated model of Climate and the Economy (RICE). These models provide comprehensive tools to understand and project the impacts of various policies and technological scenarios on carbon emissions in megacities.

Compared to the CGE model, hybrid models, and IAMs, the LEAP model is more widely adopted at the city level due to several advantages. First, it provides a comprehensive consideration of the energy system across the economic, environmental, and social domains [4]. Second, it features a flexible data structure and rich technical and end-user details [23]. Third, it has fewer requirements for the initial data and theoretical foundation, making it more accessible.

2.2.3. Socioeconomic Impacts of Low-Carbon Transformation

Some studies have explored the socioeconomic impact of the low-carbon transformation in megacities. Jasińska [27] proposed that the development of renewable energy could boost local economies by stimulating investment in rural areas, promoting regional industrial clusters, and fostering innovation and entrepreneurship. Guo [28] emphasized that a new power system will create more job opportunities in China. Zhang [29] assessed the employment impacts of Beijing’s energy transition trajectory using a soft-linking approach that combined the MESSAGEix-Beijing energy system model with IOA.

In summary, the existing literature has amassed extensive insights into the low-carbon evolution of metropolitan areas, particularly in identifying the drivers of carbon emissions and forecasting carbon footprints. However, there is a significant gap in research exploring the combined benefits that climate change mitigation and socioeconomic development could achieve through a low-carbon transformation. Addressing this gap, this study developed an integrated framework that combines the Long-range Energy Alternatives Planning (LEAP) system with a robust input–output (IO) model based on empirical data. The study aims to quantify the sector-level potential for carbon mitigation and evaluate the ripple effects of low-carbon investments on socioeconomic well-being. This approach charts a strategic course for megacities to reduce their carbon emissions while simultaneously enhancing their socioeconomic prosperity.

3. Case Study, Methodology, and Data

3.1. Case Study

This study seeks to conduct a comprehensive examination of the synergistic advantages that arise from mitigating climate change and advancing socio-economic conditions through a low-carbon transition. Thus, it has identified exemplary cases that excel in both economic prosperity and sustainable development. The report, a collaborative effort between the Chinese Academy of Social Sciences (CASS) and UN-Habitat [30], assessed the economic and sustainable competitiveness of cities on a global scale. It shows that Tokyo, New York-Newark, Los Angeles-Long Beach-Santa Ana, Paris, and Shenzhen have secured top rankings in both economic and sustainable development metrics. Moreover, 25 of the top 30 megacities globally are situated in emerging or developing economies [10]. To offer a compelling model for megacities confronting the challenges of low-carbon transformation, Shenzhen, in particular, emerges as a paradigmatic case (detailed information is provided in

Table A1. Population, economies, and sustainability of megacities.

City	Country	Population (Million)	Eco Ranking	SUS Ranking
Tokyo	Japan	37	3	1
Delhi	India	30	232	268
Shanghai	China	27	12	33
Sao Paulo	Brazil	22	429	83
Mexico city	Mexico	22	204	88
Dhaka	Bangladesh	21	350	455
Cairo	Egypt	21	385	469
Beijing	China	20	21	47
Mumbai	India	20	292	352
Osaka	Japan	19	41	10
New York-Newark	US	19	1	3
Karachi	Pakistan	16	530	500
Chongqing	China	16	196	231
Istanbul	Türkiye	15	119	76
Buenos Aires	Argentina	15	579	55
Kolkata	India	15	551	595
Lagos	Nigeria	14	464	313
Kinshasa	Congo	14	867	835
Manila	Philippines	14	258	376
Tianjin	China	14	264	114
Rio de Janeiro	Brazil	13	328	151
Guangzhou	China	13	42	69
Lahore	Pakistan	13	591	586
Moscow	Russia	13	70	12
Los Angeles-Long Beach-Santa Ana	US	12	7	23
Shenzhen	China	12	9	9
Bangalore	India	12	314	330
Paris	France	11	8	6
Bogota	Colombia	11	234	125
Chennai	India	11	383	412

Data sources: (1) United Nations, Department of Economic and Social Affairs, Population Division (New York City, USA). World Urbanization Prospects: The 2018 Revision, 2018. (2) The Chinese Academy of Social Sciences (Beijing, China); UN-Habitat (Nairobi, Kenya). Global urban competitiveness report (2020–2021): Global urban value chain: Insight into human civilization over time and space, 2021. (3) Eco Ranking and SUS ranking indicate the ranks of economic competitiveness and sustainable competitiveness, respectively.

in Appendix A).

Shenzhen is located in the Pearl River Delta in southern China. It is the core city of the Guangdong-Hong Kong-Macao Greater Bay Area, known for its high openness and economic vitality, and plays a pivotal role in China’s global competitiveness [31]. Since its designation as an economically important district in 1978, Shenzhen has experienced remarkable population and economic growth, emerging as a symbol of urban transformation in China [32]. According to the Shenzhen Statistical Yearbook [33], the permanent population at the end of the year stood at approximately 17.68 million, with the secondary and tertiary industries accounting for 37.4% and 62.5%, respectively, in 2020. Figure 1 illustrates the trends in GDP, total energy consumption, and energy intensity in Shenzhen during 2010–2021.

As illustrated in Figure 1, Shenzhen’s GDP has increased significantly, nearly tripling from approximately 1006.91 billion CNY in 2010 to 3066.49 billion CNY in 2021. The city’s total energy consumption reached 47.57 million tonnes of coal equivalent (tce) in 2021, marking a 1.46-fold increase compared with 2010, with a consistent growth trajectory, except for 2020. Consequently, Shenzhen’s energy intensity steadily decreased from 0.3246 tce per 10,000 CNY in 2010 to 0.1551 tce per 10,000 CNY in 2021. Such astonishing development may be owing to the benefits of demographics, human resources, active private economy, and technological innovation [34].

The specific accounting for carbon emissions exhibits variability due to the selection of statistical caliber and carbon dioxide emission factors. Nonetheless, the extant research provides insights into the total amount, structure, and driving factors of Shenzhen’s carbon emissions. Meng [35] employed a life cycle methodology and estimated that Shenzhen’s total carbon emissions were approximately 65.45 million tonnes in 2015. Of this total, direct carbon emissions (Scope 1) accounted for 50%, with 32.82 million tonnes, while indirect emissions totaled approximately 32.63 million tonnes. Among indirect emissions, purchased electricity consumption (Scope 2) comprised 38%, while other emissions (Scope 3) constituted 62%, including those embodied in transboundary transportation (10.54%), the upstream supply chain of critical materials (44.02%), and the downstream chain of waste disposal (7.44%).

Yu [36] reported that energy-related carbon emissions in Shenzhen were approximately 50 million tonnes in 2019, with the production and supply of electric power, steam, and hot water identified as the primary energy-intensive sector. Between 2016 and 2019, carbon emissions were distributed across sectors as follows: 0.11% in the primary sector, 21.31% in the industrial sector, 16.76% in the transport sector, 0.11% in the construction sector, and 3.7% in the service sector. Growth in GDP and population drove increased carbon emissions, while reductions in energy intensity, carbon emission factors, and upgrading of industrial structures contributed to emission reductions. In particular, between 2015 and 2019, the transport and power production sectors had a negative impact on reducing carbon emissions.

Therefore, Shenzhen, a quintessential megacity model, serves as an exemplary case for investigating the effects of its low-carbon transformation on both carbon emission reduction and socioeconomic advancement. The findings of this study will offer valuable insights into megacities, particularly those facing challenges in developing nations.

3.2. Methodology

This study integrates the bottom-up energy-environmental simulation model, LEAP–Shenzhen, with the traditional top-down economic model, the IO model, to comprehensively examine the mitigation of climate change and the socioeconomic impacts of the low-carbon transformation in the megacity of Shenzhen. The research framework is outlined in Figure 2.

3.2.1. LEAP–Shenzhen Energy System Model

The LEAP model is extensively utilized in energy policy analysis and climate change mitigation assessments [4]. Reflecting on historical trends and future projections, this study built a LEAP–Shenzhen model to delineate sector-specific carbon emissions under various low-carbon transformation strategies in Shenzhen from 2020 to 2030.

The LEAP–Shenzhen model comprises four main components. First, it establishes key assumptions regarding indicators that are strongly correlated with carbon emissions across significant sectors, such as GDP, industrial structure, population, activity level, energy consumption structure, and energy intensity. Second, it analyzes both the end-use energy demand and energy supply sides. The demand side primarily considers manufacturing, building, and transportation. The supply side incorporates five types of electricity generation: coal-fired power plants, natural gas power plants, distributed photovoltaic power plants, combined-heat- and cold-power plants, waste incineration power plants, and imports from the South Grid, which are utilized as supplementary sources when local power generation is insufficient. Third, carbon emissions in Shenzhen are calculated by integrating carbon emission factors from provincial emission inventories in China. Finally, the mitigation of climate change was evaluated under various low-carbon transformation scenarios across different sectors through scenario analysis.

The scenario analysis conducted within the LEAP–Shenzhen model projected future trends in carbon emissions under various conditions. To evaluate the effectiveness of climate-change mitigation strategies in key sectors of Shenzhen, this study established a business-as-usual (BAU) scenario as the foundational framework for developing low-carbon scenarios. The base year was set to 2020 and the simulation period spanned 2021 to 2030.

The BAU scenario operates on the premise that Shenzhen’s socioeconomic and technological progress will align with historical data and trends without the introduction of additional low-carbon strategies or technologies. This scenario highlights current and projected carbon emissions under existing policies and practices.

Based on the BAU scenario, the low-carbon scenario assumes that Shenzhen adopts additional carbon control measures and technologies. It outlines five sub-scenarios: low-carbon transformation in manufacturing, buildings, transportation, electricity generation, and a combined scenario. For each sub-scenario, feasible carbon control strategies or technologies, corresponding carbon reduction potential, and investment costs were compiled from relevant government documents, industrial surveys, and expert interviews. The low-carbon scenario aims to analyze carbon emissions in Shenzhen under various carbon control strategies. Detailed explanations are provided in

Table 1. Explanation of Scenario Construction.

Scenario	Explanation	Basis
BAU	No extra low-carbon strategies or technologies introduced.
Low carbon manufacturing	Adopting 50 specific technologies from temperature control, lighting, production, energy, control management and transportation. The promotion rate assumption is based on comprehensive consideration of unit carbon reduction cost, reduction potential, and ease of implementation.	Policy documents [33] and other relevant development plans, expert interviews, and field visits.
Low carbon building	Adopting 35 specific measures from retrofitting existing commercial and residential buildings, construction of low-emission commercial and residential buildings, and building energy management. It was assumed that the promotion rates increase linearly.	[37]
Low carbon transport	Passenger transport on road: increasing parking fees, limiting car driving, implementing HOT/HOV charging policy, encouraging “green travel” and voluntary suspensions, building a slow-travel network. Subway: Adopting 14 technologies. Freight on road: Adopting Hybrid trucks, oil-to-gas trucks, and high-efficiency fuel trucks.	Policy documents [38] and other relevant development plans, expert interviews, and field visits.
Low carbon electricity generation	Existing coal-fired power plants would be phased out; construction of photovoltaic power generation, natural gas plants and heat-power and cold-power plants; the exogenous capacity of waste incineration power plants increases.	Policy documents [39,40,41] and other relevant development plans, expert interviews, and field visits.
Combined	Considering all abovementioned low-carbon measures.

.

3.2.2. IO–Shenzhen Economic Model

IOA elucidates the interdependence among different industries by constructing IO models that reveal the level of production technology and the correlations within the economic system. This method has been widely applied in studies aimed at analyzing economic structures, forecasting economic development, and simulating policy effects [42]. In line with the actual circumstances in Shenzhen, this study employed a value-based IO table to examine the socioeconomic effects of targeted investments in low-carbon transformation.

Investments typically stimulate an increase in final output, leading to an increase in total output and subsequently generating additional demand for intermediate inputs across all sectors. In turn, this can facilitate socioeconomic improvements from various perspectives, including increased value addition, employment opportunities, and income levels. The IO table usually exhibits row, column, and aggregate equilibria. The rows indicate the output distribution of a particular sector in the urban economic system, whereas the columns indicate the input of each sector to that sector. According to Miller and Blair [43], this equilibrium is expressed as

$$X={\left(I-A\right)}^{-1}F$$

(1)

Formula (1) is the demand–pull model. Assuming that the final output, except low-carbon investment, remains constant, the impact of changes in total output, X, in each industry as a result of changes in low-carbon investment, F, can be quantitatively investigated as formula (2):

$$\Delta X={\left(I-A\right)}^{-1}\Delta F$$

(2)

This study analyzed the benefits of low-carbon transformation in socioeconomic development, focusing on increases in value-added, resident remuneration, job opportunities, and output of economic sectors.

$$G=G_0{\left(I-A\right)}^{-1}\Delta \widehat{F}$$

(3)

$$S=S_0{\left(I-A\right)}^{-1}\Delta \widehat{F}$$

(4)

$$L=L_0(I-A{)}^{-1}\Delta \widehat{F}$$

(5)

$$X=(I-A{)}^{-1}\Delta F$$

(6)

In Equations (3)–(6), G, S, L, and X denote the impacts of low-carbon investment on GDP, resident remuneration, employment, and output of the economic sectors in Shenzhen, respectively. G₀, S₀, and L₀ are the row vectors of the ratio of GDP, labor compensation, and employment at the end of the year, respectively; $∆\widehat{F}$ is the column of the composition of low-carbon investment vector. Based on the assumption that the technical coefficients do not change significantly in the short term, it is possible to quantitatively analyze the pulling effect of the change in carbon governance investment $(∆\widehat{F}$) on each indicator.

3.3. Data

3.3.1. Data of the LEAP–Shenzhen Model

This study set 2020 as the base year and 2021–2030 as the simulation year. The assumptions of the key indicators refer to historical trends and data from sources [33,44,45,46,47].

Table 2. Key indicators of the BAU scenario in the LEAP–Shenzhen model.

Indicators (Unit)	2020	2025	2030
GDP (billion CNY)	2775.90	3925.14	5252.73
Total population (10,000 persons)	2363.38	2510.00	2690.00
Proportion of secondary industry (%)	37.40	35.75	31.75
Proportion of tertiary industry (%)	62.51	64.17	68.17
Proportion of value added of manufacturing (%)	32.54	29.04	25.24
Total building area (million m2)	633.36	744.72	860.80
Total Motor vehicle ownership (vehicle)	3,337,695	4,202,333	4,938,846

lists some key indicators in the BAU scenario (see Appendix B for detailed information on the LEAP–Shenzhen model).

3.3.2. Data of the IO–Shenzhen Model

The main data on low-carbon investments come from specialized technical literature, expert interviews, and field surveys. Referring to the data published in documents such as Markaki [48] and the Reference Indicators for the Design Cost of Photovoltaic Power Generation Projects (2021 Edition) [49] issued by the Chinese government, this study disaggregates the corresponding low-carbon investments introduced in the LEAP–Shenzhen model in the decade of 2020 to 2030 into IO tables for 17 industries.

Table 3. Low-carbon investments by sector (Unit: billion CNY).

Sector	Investment	Sector	Investment
Chemical products	2.01	Wholesale and retail trade	77.73
Specialty Equipment	9.28	Transportation, storage, and postal services	24.49
Transportation equipment	7.37	Accommodation and catering	1.19
Electrical machinery and equipment	19.80	Finance	5.88
Communication equipment, computers, and other electronic equipment	20.98	Real Estate	8.56
Electricity, heat production, and supply	17.42	Leasing and business services	0.53
Gas production and supply	2.27	Integrated technical services	1.73
Water production and supply	2.28	Public administration, social security, and social organizations	2.10
Construction	258.42	Total	462.04

shows the investments in each sub-industry (see Appendix C for more information on the investments and Appendix D for more information on the industry categorization in the Shenzhen input–output tables).

4. Result and Discussion

4.1. Climate Change Mitigation Benefits from the LEAP–Shenzhen Model

This comparative analysis serves as a fundamental step in evaluating the impact of various low-carbon strategies on Shenzhen’s climate change mitigation during the simulation period (2020–2030).

Figure 3 illustrates the trend of total carbon emissions in Shenzhen from 2020 to 2030 under the BAU and combined scenarios. Under the BAU scenario, total carbon emissions in Shenzhen are projected to continue increasing without peaking during this period. Starting at 75.95 million tonnes in 2020, emissions are forecast to rise to 109.84 million tonnes by 2030. On the contrary, the combined scenario shows a different trajectory, with total carbon emissions plateauing from 2021 to 2024, peaking at approximately 82 million tonnes, followed by a decline to 76.76 million tonnes by 2030. Consequently, low-carbon initiatives have effectively altered the growth trend in carbon emissions. Moreover, such efforts align with Shenzhen’s goal of achieving peak carbon emissions before 2030, as outlined in the Shenzhen Carbon Peak Implementation Plan of the Shenzhen Municipal Government for 2023 [50].

A comparison of Shenzhen’s carbon-peaking pathways with those of Beijing and Shanghai revealed distinct trends. In Beijing, carbon emissions peaked in 2011 at 94.4 million tonnes but experienced a subsequent increase from 2017 to 2019, increasing from 85 million tonnes to 88.15 million tonnes. However, with the adoption of feasible carbon control measures, emissions could potentially decrease to approximately 65 million tonnes by 2030 [51]. In Shanghai, carbon emissions reached 193 million tonnes in 2019 without peaking. However, low-carbon measures can facilitate a peak in 2025 at ~240.9 million tonnes, followed by a decrease to 85.26 million tonnes by 2060 [52]. These comparisons underscore the importance of effective low-carbon transformation strategies for achieving emission reduction targets and transitioning toward sustainable development in megacities such as Beijing, Shanghai, and Shenzhen.

Figure 4 illustrates the comparisons of carbon intensity and carbon emission per capita in Shenzhen from 2020 to 2030 under the BAU and combined scenarios. Carbon intensity was approximately 0.27 tonnes per ten thousand CNY in 2020, peaking at 0.29 in 2021 and decreasing to 0.21 in 2030 under the BAU scenario. Under the combined scenario, carbon intensity continued to decline, reaching 0.16 tonnes per ten thousand CNY in 2030, representing a decrease rate of 23.81%. These results indicate that, despite carbon intensity peaking in 2021 without external carbon control measures, the adoption of carbon control technologies or strategies can significantly reduce carbon intensity.

Carbon emissions were approximately 4.31 tonnes per capita in 2020 and increased to 5.51 in 2030 without peaking under the BAU scenario. In contrast, carbon emissions peaked at approximately 4.63 in 2021 and 2022 and then continuously decreased to 3.85 in 2030 under the combined scenario, with a decrease rate of 30.13%. The results showed that low-carbon measures have the potential to reverse the continuous growth trend of carbon emissions per capita, peaking in 2022.

Compared to total carbon emissions, carbon intensity, and carbon emissions per capita, the LEAP–Shenzhen model indicates that carbon intensity peaks first, followed by carbon emissions per capita, and total carbon emissions gradually peak under carbon control measures. This observation aligns with the findings of Dong [53], who divided the carbon peaking process into three stages based on developed economies. Additionally, Dong [53] proposed that the level of industrialization will continue to increase before the peak of total carbon emissions, a trend consistent with the trajectory observed in Shenzhen.

Figure 5 illustrates the changes in the carbon emission structures from 2020 to 2030 under the BAU and combined scenarios. Figure 5a shows that in 2020, the electricity generation and transport sectors were the major carbon emitters, contributing 63% and 30% of emissions, respectively. Figure 5b shows that under the BAU scenario, the proportion of carbon emissions from electricity generation increases to 76%, while emissions from the transport sector decrease to 18% by 2030. Conversely, under the combined scenario, electricity generation accounted for 72% of total emissions, whereas the transport sector accounted for 21% in 2030.

The IEA report [54] revealed that electricity/heat generation and transport contributed to two-thirds of total CO₂ emissions and have been responsible for almost the entire global growth in emissions since 2010, with the remaining third split between industry and buildings [55]. The carbon emissions structure from the LEAP–Shenzhen model closely aligns with this; however, electricity/heat generation and transport hold a higher proportion. This is because Shenzhen must purchase a large amount of electricity from outside and retain a significant number of fuel-powered vehicles, despite the vigorous promotion of new energy vehicles. Shanghai faces a similar situation, in which CO₂ emissions from electricity remain high and transportation is the main sector of urban energy consumption [52]. Similarly, in Beijing, power and transportation contribute significantly to carbon emission reduction efforts [51]. These results highlight the importance of low-carbon transformations in megacities, particularly focusing on the power and transportation industries, such as improving the cleanliness of electricity, developing new energy vehicles and public transportation, and restricting private car usage.

4.2. Socioeconomic Benefits from the IO–Shenzhen Model

The total low-carbon investment of 462.04 billion CNY was divided into 17 sectors (Table 3). Using the Shenzhen IO table for 2020 as a base, the socioeconomic benefits of the low-carbon transformation were evaluated, focusing on total output, GDP, resident remuneration, and employment across sectors (Figure 6).

From the perspective of economic benefits, total output can increase by 799.490 billion CNY with an impact multiplier of 1:1.73, while GDP can increase by 311.416 billion CNY with an impact multiplier of 1:0.67. A detailed analysis of the output effect reveals that it primarily affects the construction industry, wholesale and retail trade, financial services, communication equipment, computers, transportation, warehousing, and postal services. These five industries collectively account for approximately 85.62% of the output influenced by these investments. Moreover, low-carbon investment also has a substantial effect on GDP. According to the IO–Shenzhen model, the construction industry contributes the most to the growth of GDP stimulated by carbon governance investments, which amounts to 153.702 billion CNY, followed by wholesale and retail trade, transportation, storage, postal services, communication equipment, computers, and real estate, which collectively made up 42.86% of the total.

From the perspective of social benefits, resident remuneration increased by 156.10 billion CNY, with an impact multiplier of 1:0.34 over the decade. Additionally, it can create 1.79 million job opportunities, with every 100 million CNY of low-carbon investment resulting in 387 job opportunities. Income and employment affect flow mainly to the following sectors: (1) construction industry; (2) wholesale and retail trade; (3) electrical machinery and equipment; (4) transportation, warehousing, and postal services; (5) communications equipment, computers; and (6) other electronic equipment.

4.3. Efficiency of Low-Carbon Investment

The key results of the LEAP–Shenzhen and IO–Shenzhen models were evaluated to assess the efficiency of low-carbon investments. Such an analysis offers valuable insights for policymakers and researchers to advance carbon peak and carbon neutrality initiatives while simultaneously achieving sustainable development goals.

Table 4. Efficiency of low-carbon investments.

Sectors	Carbon Reduction (Million tonnes/Billion CNY)	Employment (Person/Billion CNY)	Output per Unit of Investment	Value-Add per Unit of Investment	Resident Remuneration per Unit of Investment
All	0.2446	3946.3411	1.7252	0.6727	0.3403
Manufacturing	1.5121	2086.7168	1.9171	0.4617	0.2734
Building	0.1554	4598.3482	1.6171	0.7448	0.3738
Transport	0.0418	3623.6407	1.8504	0.5948	0.3076
Electricity generation	1.1065	1754.8846	1.8062	0.7067	0.2874

provides detailed information.

As illustrated in Table 4, considering all feasible low-carbon strategies, a one-billion -CNY investment has the potential to reduce carbon emissions by 0.2446 million tonnes and generate 3946.3411 job opportunities in Shenzhen. On a per-CNY investment basis, this could stimulate 1.7252 and 0.6727 CNY increases in output and value-added, respectively, and a 0.3403 CNY increase in resident remuneration during policy implementation.

Compared across sectors, low-carbon investments in manufacturing exhibit the highest efficiency in reducing carbon emissions. A one-billion-CNY investment can reduce carbon emissions by 1.5121 million tonnes in manufacturing, nearly 10 times that in buildings, and 36 times that in transport. Following manufacturing, electricity generation shows an elasticity of 1.1065, which is 7.12 times higher than that of building and 26.47 times that of transport.

In terms of socioeconomic progress, low-carbon investments in the building and transportation sectors demonstrate relatively higher efficiency in terms of job creation and increasing resident remuneration. A one-billion-CNY investment in the building sector can create 4598.3482 job opportunities, 2.2 times that in manufacturing and 2.6 times that in electricity generation. Similarly, a one-billion-CNY investment in the transport sector can create 3623.6407 job opportunities, 1.7 times that in manufacturing, and 2.06 times that in electricity generation. The elasticity of resident remuneration in the building sector was 0.3076, 11.12% higher than that in manufacturing and 6.55% higher than that in electricity generation. In the transportation sector, the elasticity of resident remuneration is 0.3738, 26.87% higher than that in manufacturing and 23.11% higher than that in electricity generation. The value-added effect is more efficient in the building and electricity generation sectors, with elasticities of 0.7448 and 0.7067, respectively.

Therefore, from the perspective of capital utilization efficiency, investing in manufacturing and electricity generation can effectively reduce carbon emissions. Conversely, investing in the building and transportation sectors demonstrates higher efficiency in socio-economic development. Given that transportation and electricity generation are key contributors to emissions in Shenzhen, the government should prioritize improving capital utilization efficiency in the transportation sector.

5. Conclusions and Policy Implications

There is increasing recognition of the damage caused by climate change and the significant role that megacities play in mitigating climate change and fostering socioeconomic progress. This study investigated the effects of low-carbon transformation in a Chinese megacity, Shenzhen, by integrating the bottom-up energy-economic model, LEAP, with the top-down economic model, input–output analysis.

The conclusions of this study are as follows. First, low-carbon transformation measures can significantly mitigate climate change. The LEAP–Shenzhen model indicates that carbon emissions will continue to increase until 2030; however, feasible low-carbon strategies could cause a peak before 2025, resulting in a total reduction of 195.66 million tons of carbon emissions from 2020 to 2030. Carbon peaking follows a mechanism in which carbon intensity peaks earlier, followed by carbon emissions per capita and then total carbon emissions. Second, low-carbon transformation not only benefits climate change mitigation but also benefits socioeconomic advancement. The input–output-Shenzhen model demonstrates that a total investment of 462.04 billion CNY in low-carbon transformation could potentially yield approximately 799.49 billion CNY in output, 311.42 billion CNY in value-added, and 156.10 billion CNY in resident remuneration and create 1.79 million job opportunities throughout the period. Construction, wholesale, and retail have emerged as the main beneficiaries. Finally, from an efficiency perspective, low-carbon investments in the manufacturing and electricity generation sectors perform better in mitigating climate change, while those in buildings and transportation perform better in promoting socioeconomic development. These findings highlight the potential to achieve synergy between addressing climate change and promoting socioeconomic development in megacities.

Based on these conclusions, this study proposes the following policy recommendations to assist Shenzhen in achieving the peak of carbon and socioeconomic development, thus serving as a model for other megacities. First, it focuses on prioritizing and improving the cleanliness of electricity, particularly imported electricity. Electricity consumption has been increasing with the acceleration of electrification in various sectors. In 2020, electricity accounted for almost 63% of total emissions in Shenzhen. Efficiency analysis indicated that low-carbon investment in electricity generation can significantly contribute to climate change mitigation, with a 1 billion investment potentially reducing carbon emissions by 1.1065 million tonnes. Therefore, it is advisable to prioritize low-carbon initiatives that aim to improve the cleanliness of electricity. This could involve expanding renewable electricity generation, both locally and through imports, and establishing a green power market to incentivize the development of renewable energy sources. Second, it improves the efficiency of low-carbon investments in the transportation sector. Transportation is a significant contributor to carbon emissions, accounting for nearly 30% of total emissions in Shenzhen. However, although low-carbon investment in transportation may perform well in stimulating socioeconomic indicators, it does not demonstrate the same effectiveness in reducing carbon emissions. A one-billion-CNY investment in transport carbon governance results in only a modest reduction of 0.0418 million tonnes of carbon emissions. Therefore, it is necessary to explore more efficient carbon control policies in the transportation sector. This study provides valuable insights for policymakers, emphasizing the importance of considering both climate change mitigation and socioeconomic effects when designing carbon-peaking strategies.

Although this research contributes to the existing literature on low-carbon transformation in megacities by comprehensively evaluating its benefits not only in mitigation of climate change but also in socioeconomic advancement, it has three main limitations. (1) This study draws on Shenzhen’s extensive adoption of low-carbon strategies since 2015 and summarizes feasible measures from government documents, expert interviews, and field investigations. However, it may have limitations, such as subjectivity and bias in scenario analysis. (2) Although this study provides a potential plan for megacities to achieve carbon peaks alongside socioeconomic improvements, its representativeness is somewhat limited. Megacities at different stages of industrialization and urbanization usually face distinct conditions that may require tailored approaches. Addressing these limitations in future studies could enhance the robustness and applicability of the research findings, enabling more effective and tailored carbon governance strategies for megacities around the world. (3) While input–output analysis is grounded in a robust theoretical framework, its dependency on fixed coefficients within specific input–output tables constrains the flexibility of the analysis. The application of machine learning provides a data-driven approach to dynamically adjust and optimize these coefficients. Therefore, it is advisable to explore machine learning techniques to enhance the adaptability of input–output analysis, particularly in mitigating the limitations posed by fixed coefficients in the future.