The Carbon Reduction Contribution of Battery Electric Vehicles: Evidence from China

Abstract

The transition to passenger car electrification is a crucial step in China’s strategic efforts to achieve carbon peak and carbon neutrality. However, previous research has not considered the variances in vehicle models. Hence, this study aims to fill this gap by comparing the carbon emission reduction and economic feasibility of battery electric vehicles (BEVs) in the Chinese market, taking into account different powertrains, vehicle segments, classes, and driving ranges. Next, the study identifies the most cost-effective BEV within each market segment, employing life-cycle assessment and life cycle cost analysis methods. Moreover, at different levels of technological development, we construct three low-carbon measures, including electricity decarbonization (ED), energy efficiency improvement (EEI), and vehicle lightweight (LW), to quantify the emission mitigation potentials from different carbon reduction pathways. The findings indicate that BEVs achieve an average carbon reduction of about 31.85% compared to internal combustion engine vehicles (ICEVs), demonstrating a significant advantage in carbon reduction. However, BEVs are not economically competitive. The total life cycle cost of BEVs is 1.04–1.68 times higher than that of ICEVs, with infrastructure costs accounting for 18.8–57.8% of the vehicle’ s life cycle costs. In terms of cost-effectiveness, different models yield different results, with sedans generally outperforming sport utility vehicles (SUVs). Among sedans, both A-class and B-class sedans have already reached a point of cost-effectiveness, with the BEV400 emerging as the optimal choice. In low-carbon emission reduction scenarios, BEVs could achieve carbon reduction potentials of up to 45.3%, 14.9%, and 9.0% in the ED, EEI, and LW scenarios, respectively. Thus, electricity decarbonization exhibits the highest potential for mitigating carbon emissions, followed by energy efficiency improvement and vehicle lightweight. There are obvious differences in the stages of impact among different measures. The ED measure primarily impacts the waste treatment process (WTP) stage, followed by the vehicle cycle, while the EEI measure only affects the WTP stage. The LW measure has a complex impact on emission reductions, as the carbon reductions achieved in the WTP stage are partially offset by the increased carbon emissions in the vehicle cycle.

1. Introduction

In 2022, China’s total carbon emissions reached about 12.1 billion tons, constituting 28.87% of the global total [1]. The transportation sector emerges as the third-largest contributor to carbon emissions in China, with passenger cars accounting for 58% [2]. Despite being the world’s largest automotive market, China still exhibits a relatively low automobile ownership per thousand people compared to developed countries, indicating a greater potential for the growth of the automobile market. For example, China had approximately 226 cars per thousand people in 2022, while the United States had around 837 cars per thousand people. This rapid surge in vehicle demand would lead to a considerable increase in petroleum consumption and air pollution. To mitigate environmental impacts and restrain the growth of energy demand, China has committed to achieving carbon peaking by 2030 and carbon neutrality by 2060 [3]. The automobile industry is therefore confronted with a structural transformation challenge in the form of replacing internal combustion engines (ICEVs) with low-carbon alternatives that can meet environment-friendly standards. Among these alternative vehicles, battery electric vehicles (BEVs) have been proposed as one of the most promising alternatives to ICEVs, owing to their advantages in terms of high efficiency, low emissions, and zero-tailpipe emissions.

The high purchase costs and inadequacy of charging infrastructure present significant challenges to the widespread penetration of BEVs in China. The Chinese government has implemented a series of policies including production subsidies, consumer incentives, and subsidies for charging infrastructure construction to accelerate the penetration rate of BEVs [4]. In response, automotive manufacturers diversify the types and driving ranges of BEVs to align with government standards and meet market requirements, thereby enhancing their competitiveness. Additionally, fuel cell vehicles (FCVs) are also emerging as a potential solution in the pursuit of carbon neutrality. FCVs use hydrogen as fuel and produce only water as a by-product during operation, offering a truly zero-emission driving experience. In the current global decarbonization trend, the vigorous promotion of renewable energy and green hydrogen provides opportunities for the rapid development of FCVs in the future. Renewable energy can be used to produce green hydrogen through electrolysis, which can then power FCVs, creating a sustainable energy cycle. However, like BEVs, FCVs also face challenges such as high costs and insufficient refueling infrastructure [5]. Although FCVs are not the main focus of this study, it is important to note their potential in the automotive decarbonization landscape.

However, there are many BEVs available with variations in segments, classes, and driving ranges in the Chinese market. Each type of BEV exhibits different characteristics, including energy consumption, greenhouse gas emissions, initial purchase costs, energy costs, and emission costs [6]. This diversity prompts questions about whether all types of BEVs meet both environmental and economic viability standards. Automobile manufacturers encounter challenges when determining the most effective type of battery passenger car to invest in. Hence, it is necessary to conduct a comprehensive analysis of the carbon reduction potential and economic viability of battery passenger cars across various segments and identify the most cost-effective powertrain system, vehicle segment, class, and driving range. Such an in-depth analysis is crucial for guiding decision-making in the pursuit of sustainable and economically viable electric vehicles.

Several previous studies have extensively discussed the carbon emissions of new energy vehicles (NEVs). Most studies consistently demonstrate that NEVs have greater advantages in energy savings and emissions reduction compared to ICEVs. For example, the greenhouse gas (GHG) emissions of BEVs are 47% lower than those of ICEVs [7]. This difference is primarily concentrated in the vehicle usage stage, although the carbon emissions associated with the production, assembly, distribution, and recycling of the BEVs can partly offset the reduced emissions during usage. By combining the fuel and vehicle cycles and considering battery recycling and disposal, results show BEVs can reduce GHG emissions by 18% compared to ICEVs [8]. With proper battery recycling, BEVs can achieve more environmental benefits. Additionally, BEVs with different lithium-ion batteries can reduce emissions by 18.3–22.6% compared to ICEVs [9].

However, differences in the selection of system boundaries, functional units, and vehicle segments have led to diverse results, as listed in Table 1.

Table 1. The representative literature on GHG emissions of passenger cars.

Author	Model Year	Region	Powertrain System	Functional Units (Years; km)	Vehicle Segment	Electric Range (km)
Yang et al. (2021) [10]	2019	China	ICEV; BEV; PHEV	10 150,000	A-class sedan	358
Li et al. (2023) [11]	2021	China	MV; BEV; ICEV	/ 600,000	A-class sedan	421
Luo et al. (2022) [12]	2020	China	MV; CNGV; BEV; ICEV	8 250,000	B-class sedan	606
Xiong et al. (2019) [13]	2018	China	BEV; PHEV	/ LFP: 160,000 NMC: 120,000	A-class sedan	300/450
Kannangara et al. (2021) [14]	2020	Canada	ICEV; HEV; PHEV; BEV; FCEV	13 200,000	A0-class sedan	295
Joshi et al. (2022) [15]	2021	Nepal	BEV; FCEV; ICEV	20 200,000	A0-class SUV	500

PHEV: plug-in hybrid electric vehicle; MV: motor vehicle; CNGV: compressed natural gas vehicle; HEV: hybrid electric vehicle; FCEV: fuel cell electric vehicle; LFP: lithium iron phosphate; NMC: nickel manganese cobalt.

Table 1. The representative literature on GHG emissions of passenger cars.

Author	Model Year	Region	Powertrain System	Functional Units (Years; km)	Vehicle Segment	Electric Range (km)
Yang et al. (2021) [10]	2019	China	ICEV; BEV; PHEV	10 150,000	A-class sedan	358
Li et al. (2023) [11]	2021	China	MV; BEV; ICEV	/ 600,000	A-class sedan	421
Luo et al. (2022) [12]	2020	China	MV; CNGV; BEV; ICEV	8 250,000	B-class sedan	606
Xiong et al. (2019) [13]	2018	China	BEV; PHEV	/ LFP: 160,000 NMC: 120,000	A-class sedan	300/450
Kannangara et al. (2021) [14]	2020	Canada	ICEV; HEV; PHEV; BEV; FCEV	13 200,000	A0-class sedan	295
Joshi et al. (2022) [15]	2021	Nepal	BEV; FCEV; ICEV	20 200,000	A0-class SUV	500

PHEV: plug-in hybrid electric vehicle; MV: motor vehicle; CNGV: compressed natural gas vehicle; HEV: hybrid electric vehicle; FCEV: fuel cell electric vehicle; LFP: lithium iron phosphate; NMC: nickel manganese cobalt.

However, it should be noted that the adoption of NEVs may result in increased particulate matter emissions. A comparative study on the environmental impact of ICEVs and NEVs across 34 major cities in China revealed that although certain emissions from NEVs were similar to those from ICEVs, specific particulate emissions were 19 times higher than those generated by ICEVs [16]. Similarly, it was observed that while NEVs exhibit significant reductions in some pollutants compared to ICEVs, they also demonstrate an increase in emissions of others [10]. This discrepancy highlights the complexity of evaluating the environmental benefits of NEVs, as reductions in GHG emissions may be offset by harmful increases in particulate matter. Therefore, it is crucial to quantify the costs associated with pollutant emissions from a societal perspective to better understand the socio-economic impact of environmental pollution caused by the adoption of NEVs. This holistic approach should consider the health and economic effects of increased particulate matter in addition to the benefits of lower GHG emissions.

Compared to China, numerous developed countries show a greater potential for energy-saving and emission reduction in the field of NEVs. The primary reason contributing to this disparity lies in China’s heavy dependence on coal and other fossil fuels in its electricity structure, resulting in higher GHG emission intensity. An analysis highlighted notable differences in carbon emissions among nations, influenced by their respective electricity structures. For example, in developing countries where coal-fired power generation remains predominant, the emissions from NEVs are much higher than in regions with predominantly renewable energy sources [9]. Regional variations in the carbon reduction potential of BEVs have also been identified, driven by factors such as renewable portfolio standards and electricity structures [17]. Additionally, it has been emphasized that the development of NEVs must be accompanied by decarbonization of the power sector [18]. These findings underscore the importance of optimizing electricity structures to accelerate the adoption of NEVs.

However, influencing factors extend beyond cleaner energy transitions, with technological advancements playing an equally critical role in enhancing the carbon reduction potential of vehicles. The promoting effect of technological progress on carbon reduction throughout the entire life cycle of vehicles has been widely verified. Studies show that lightweighting is an effective way to reduce energy consumption and carbon emissions. For every 100 kg decrease in the weight of a car, the fuel consumption per 100 km can be reduced by 0.4 L, and the carbon emissions can be reduced by approximately 1 kg. Data from the U.S. Department of Energy also shows that for every 10% reduction in a car’s weight, its fuel consumption decreases by 6% to 8% [19]. Reducing the curb weight of automobiles can effectively reduce energy consumption. New energy vehicles have an urgent demand for lightweighting [20]. Lithium-ion batteries, as the mainstream power battery technology, have seen their energy density increase from the early 100–150 Wh/kg to around 250 Wh/kg, and their driving range has also risen from the initial 100 km to over 400 km. Meanwhile, new battery technologies such as solid-state batteries, lithium–sulfur batteries, and sodium-ion batteries are also constantly making breakthroughs, enhancing energy efficiency and driving range while also promoting carbon reduction [21]. Research reports from institutions such as the International Energy Agency (IEA) show that with technological advancements in engine technology, electric drive systems, and other areas, the energy efficiency of various vehicles has been continuously improving, and carbon emissions per unit of mileage have been continuously decreasing [22]. It can be seen that technological progress will also promote carbon reduction by improving the energy efficiency of vehicles.

The widespread adoption of BEVs is highly dependent on consumer behavior and preferences. One of the pivotal factors from a consumer’s standpoint is the life cycle cost (LCC) of BEVs. Currently, the total cost of owning an electric passenger car is higher than that of ICEVs, primarily due to the initial purchase price [23]. To further investigate the economic viability of NEVs, a study considered consumer heterogeneity, quantitative range anxiety cost, and monetized tangible and intangible policies and developed a Monte Carlo simulation-based analysis model for assessing total cost of ownership. The research findings indicated that BEVs are the most economical choice in cities with purchase restrictions; however, ICEVs still dominate the market in cities without purchase restrictions [24]. From a consumer perspective, it was found that BEVs lack competitiveness compared to ICEVs and require national and local subsidies to offset the cost gap in the short term [25]. However, another study discovered that BEVs become more competitive than ICEVs when there are incentives available, high vehicle usage, or larger vehicle sizes when considering the long-term ownership costs [26]. Additionally, an assessment of BEVs’ economic competitiveness from consumer, societal, and GHG emissions perspectives demonstrated that even with central government subsidies, BEVs will not be cost-effective unless the annual external cost reduction reaches $2500 for a compact vehicle or $3600 for a multi-purpose vehicle [27].

The life cycle assessment (LCA) or life cycle cost alone does not fully reflect the market competitiveness and sustainability characteristics of NEVs, and a cost-effectiveness analysis (CEA) is therefore necessary to justify whether a vehicle meets both environmental and economic feasibility standards. The early studies concluded that NEVs were not cost-effective, which was reasonable considering the immature technological development level of NEVs at that time [28]. Although the CEA framework of the research is still convincing, the market for alternative fuel vehicles (AFVs) has changed considerably since the original research was carried out. Only a few studies have provided updated insights into the cost-effectiveness of electric vehicles, taking into account advancements in technology and changing market conditions. Hybrid electric vehicles (HEVs) were found cost-effective, and BEVs could be cost-effective with increases in gasoline prices or improvements in battery learning rates. Moreover, the cost-effectiveness of BEVs diminishes as the driving range increases [29]. Optimal cost-effectiveness was achieved when the gasoline price reached 9.8 CNY/L [30]. And a CEA on fuel cell vehicles (FCVs) and BEVs was conducted in South Korea and found that FCVs would need a 72.3% reduction in purchase costs to achieve cost-effectiveness, while BEVs would require a 45.3% reduction [31].

In summary, current research still shows several shortcomings: Most studies only focus on the one-dimensional analysis of the life cycle carbon emissions or life cycle costs, and the selected vehicle segments and classes are not the same, resulting in a lack of accuracy and comparability in third-party comparisons. Moreover, many existing studies were centered on specific vehicle models, such as an A-class vehicle, which may not fully represent the performance of vehicle models available in the automobile market. Hence, we select 37 real-world vehicle models with consideration of different vehicle segments, classes, and electric driving ranges for the two-dimensional analysis of cost-effectiveness. Additionally, only a few studies have paid attention to the infrastructure costs and environmental costs in the total life cycle costs of vehicles. Thus, we take into account the infrastructure costs of considering different charging methods (home charging, public AC charging, and public DC charging) and calculate the environmental costs using environmental taxes. Finally, in addition to the electricity mix, there are still many other factors that have an impact on the carbon reduction potential of BEVs. Therefore, we discuss the impact of other low-carbon measures on the emission reductions potential, such as the improvement in energy efficiency and vehicle lightweight degree, and identify the measures with the highest emission mitigation potential.

The goal of this study is to determine the most cost-effective powertrain system, vehicle segment, class, and driving range for passenger cars when fully considering charging infrastructure costs and environmental emission costs. This paper is one of the initial comprehensive cost-effectiveness analyses of the Chinese market that considers various vehicle types. The potential contributions of this study are as follows: (1) Broaden the research scope to encompass a comprehensive evaluation of the life cycle assessment, life cycle cost, and cost-effectiveness of both ICEVs and BEVs. The study aims to provide a more holistic assessment of the different powertrain systems across segments by considering factors such as vehicle segments, classes, and electric driving ranges. (2) From a cost perspective, the study incorporates the costs of charging infrastructures, environmental impacts and energy costs considering different charging habits into the life cycle cost boundary. (3) Furthermore, this study provides an in-depth exploration of influencing factors to forecast carbon reduction potentials. Considering different technological development levels (low, medium, high), three low-carbon measures are established for the years 2025, 2030, and 2035. These measures include electricity decarbonization (ED), energy efficiency improvement (EEI), and vehicle lightweight (LW).

The structure of this paper is organized as follows: Section 2 introduces the methods, data, and scenarios used in this study. In Section 3, we discuss the results derived from the life cycle assessment (LCA), life cycle cost (LCC) analysis, cost-effectiveness model, and scenario analysis, and we introduce the emission reduction potential in three low-carbon emission reduction scenarios. Finally, Section 4 provides conclusions and policy implications and outlines potential future work directions.

2. Materials and Methods

2.1. Research Framework

The framework of this research is shown in Figure 1. In this study, we compare the cost-effectiveness of ICEVs and BEVs to determine the most cost-effective vehicles. The scope of this research encompasses life cycle assessment, life cycle cost, cost-effectiveness, and scenario analysis. The LCA consists of two phases: the fuel cycle and the vehicle cycle. The fuel cycle encompasses the well-to-pump of fuel production, transportation, storage, refueling, waste treatment process (WTP), and pump-to-wheel of vehicle operations (PTW). Meanwhile, the vehicle cycle includes vehicle manufacture, assembly, distribution, maintenance, and end-of-life processes (like scrapping and disposal). The model examines six components contributing to the life cycle cost of a vehicle: purchase cost ($PC$), charging infrastructure cost ($IC$), energy cost ($EC$), emission cost ($SC$), maintenance cost ($MC$), and scrap value ($SV$). Additionally, considering different levels of technological development, we construct three low-carbon measures, including electricity decarbonization, energy efficiency improvement, and vehicle lightweight, to evaluate the carbon reduction potential of different paths on the battery electric vehicles (BEVs).

2.2. Vehicle Model Selection

Passenger cars occupy a substantial portion of the Chinese automotive market, making them the focal point of this study. Passenger cars can be classified into sedans, sport utility vehicles (SUVs), and multi-purpose vehicles (MPVs). In 2022, sedans and SUVs accounted for 95.4% of the passenger car market. Therefore, this study predominantly focuses on these two segments to ensure the research’s relevance and representativeness.

From the perspective of technical specifications, vehicles are classified into ICEVs and BEVs based on the type of power system, that is, the energy source and power conversion method relied upon for vehicle drive. In terms of vehicle model grades, according to the commonly used classification standards for automotive grades, sedans and SUVs are further sub-divided into different classes: A00-, A0-, A-, B-, C-, and D-class sedans and A0-, A-, B-, C- and D-class SUVs, according to the commonly used classification standards for car grades. Among them, D-class cars, due to their high prices and high-end positioning, mainly target a few high-consumer groups and have a relatively low frequency of appearance in daily life [32]. Therefore, D-class models were not included in this study. These class differences allow for a more detailed analysis of the electric passenger car market in China. Furthermore, electric passenger cars vary in their electric driving ranges. To cover this range of electric driving ranges, this study set 4 reference driving ranges at 100 km intervals, ranging from 200 km to 500 km [33].

Regarding the selection of vehicle models, we based our choice primarily on market share and technical specifications, including powertrain systems, vehicle classes, and driving ranges to ensure the studied models comprehensively represent mainstream market offerings while maintaining analytical rigor. Thus, we selected 37 real-world vehicle models as research objects, taking into account different powertrain systems, vehicle segments, classes, and driving ranges. These selected vehicles are the most representative models of the mainstream vehicles available for sale on the market in 2022, and have similar body sizes within the same segment. The detailed parameters of these models sourced from the Autohome website are listed in

Table 2. Parameters of representative ICEVs models.

Segment	Class	Curb Weight (kg)	Fuel Consumption (L/100 km)
Sedan	A00	930	5.7
	A0	1108	5.57
	A	1258	5.94
	B	1479	6.6
	C	1804	8.9
SUV	A0	1108	5.57
	A	1545	6.39
	B	1605	6.99
	C	2005	7.89

and

Table 3. Parameters of representative BEVs models.

Segment	Class	Driving Range (km)	Curb Weight (kg)	Battery Capacity (kWh)	Energy Density (Wh/kg)	Power Consumption (kWh/100 km)
Sedan	A00	200	772	17.3	115	9
		301	975	29.2	140.01	9.4
		403	1190	41	130.09	11
	A0	305	1058	29.9	125.5	10.6
		401	1450	44.928	140.56	11.3
		501	1510	59.1	177.2	13.2
	A	300	1260	30.7	146	11.5
		400	1580	47.5	140	12
		500	1650	57	140	12.3
	B	330	1620	49.7	140.5	17.1
		480	1950	60.2	125	13.8
		526	2029	70	138	14.3
	C	300	1900	54.6	126.35	18.2
		495	2360	83.7	130	20.1
		530	2349	75	142.1	16.2
SUV	A0	201	921	15.97	110	9.3
		303	1006	28.1	130	11
		401	1151	38.54	160	11.5
		535	1770	66	184.4	13.4
	A	344	1890	52.5	143	16
		425	1960	57.3	155	14
		505	2040	71.8	141	15.3
	B	308	1950	49	115.32	16.3
		400	2030	65.6	182.3	15.5
		500	1929	60	126	12.7
	C	355	2460	70	135.65	19
		485	2361	75	142.1	17.6
		500	2280	90.38	180	18.03

Data source: https://www.autohome.com.cn/beijing/ (accessed on 27 January 2025).

.

2.3. Model and Scenario Setting

2.3.1. Life Cycle Assessment

BEVs are generally recognized as environmentally friendly because of their low GHG emissions, emitting zero pollutants while in operation. Nevertheless, the process of producing electricity results in a considerable amount of GHGs. For this reason, the environmental impact of BEVs should be evaluated from a cradle-to-grave perspective that takes into account all stages from raw material extraction to disposal. We modified the GREET version 2022 based on real Chinese data and production processes to calculate life cycle energy consumption and GHG emission, which was developed by Argonne National Laboratory [34]. In 2022, China’s electricity structure remained highly dependent on coal, accounting for 58.40%; nuclear power accounted for 4.72%, and other energy sources accounted for 36.88%.

The lifetime driving distance of 150,000 km [35] and an average lifespan of 10 years are considered as the functional units for our analysis. Although there are NMC lithium-ion batteries with different stoichiometric ratios available in the market, this study focuses on the NMC 811 battery as the reference, and no battery replacement will be required. Thermo-metallurgical and hydrometallurgical technologies are two major vehicle recycling technologies globally. We assume that hydrometallurgical technology for vehicle recycling will be extensively used in the future, and the lifetime GHG emissions of the charging infrastructure should be ignored for the reason that they only account for a small percentage [36].

2.3.2. Life Cycle Cost

Because the transition to AFVs brings about environmental benefits that contribute to societal well-being, the costs to be considered should be those borne by the whole of society. We therefore take into account the costs incurred by governments, consumers, and businesses in switching to AFVs and compare these costs with those associated with the existing ICEV system. Life cycle cost (LCC) refers to the present value of all the costs associated with a vehicle during its service lifetime. We consider the cost of charging infrastructure in the LCC. The environmental cost is also included in the research framework from a social benefit perspective.

$PC$ is a one-time input, which is the present value, while future costs such as $EC$, $SC$, $MC,$ and $SV$ need to be converted into the present value to keep comparability. The total life cycle cost of a vehicle can be expressed as

$$TC=PC+IC+\sum _{k=1}^n({\displaystyle \frac{EC+SC+MC}{(1+r{)}^{k-1}}}-{\displaystyle \frac{SV}{(1+r{)}^{n-1}}})$$

(1)

where $n$ is the lifetime of the vehicle, and $r$ is the discount rate. The recommended social discount rate for short- and medium-term projects in China is 8%, while for long-term projects, it is less than 8% [37]. Some studies assume that the discount rate for battery electric vehicles or fuel vehicles throughout their entire life cycle is 6% [38]. Therefore, we assume that $n$ is 10 years, and $r$ is 6% in this analysis.

1.

Purchase cost

The purchase cost of a vehicle includes the manufacturer’s suggested retail price $\left(MSRP\right)$, the purchase tax $\left(VPT\right)$, the vehicle and vessel use tax $\left(VUT\right)$, the license fee $\left(LF\right)$, and the government subsidy $\left(GS\right)$. The purchase cost can be expressed by

$$PC=MSRP+VPT+VUT+LF-GS$$

(2)

where $MSRP$ is sourced from the Autohome website, where the manufacturer’s recommended price is published. According to the “Vehicle Purchase Tax Law of the People’s Republic of China”, the $VPT$ for ICEVs is 10% of the $MSRP$, while BEVs are exempt from this tax, resulting in ${VPT}_{BEV}=0$. Furthermore, the $VUT$ is a local tax that may vary significantly between provinces and cities. In this study, the $\mathrm{V}\mathrm{U}\mathrm{T}\mathrm{s}$ for provincial capital cities are used as a representation for provincial taxes. There are different vehicle and vessel use taxes in different provinces (see

Table 4. Vehicle and vessel tax standards across provinces (CNY/year).

	$\mathit{D} < 1.0$	$1.0 \le \mathit{D} < 1.6$	$1.6 \le \mathit{D} < 2.0$	$2.0 \le \mathit{D} < 2.5$	$2.5 \le \mathit{D} < 3.0$	$3.0 \le \mathit{D} < 4.0$	$\mathit{D} \ge 4.0$
Beijing	300	420	480	900	1920	3480	5280
Shanghai	180	360	450	720	1500	3000	4500
Guangdong	180	360	420	720	1800	3000	4500
Zhejiang	180	300	360	660	1500	3000	4500
Tianjin	270	390	450	900	1800	3000	4500
Chongqing	120	300	360	660	1200	2400	3600
Jiangsu	120	300	360	660	1200	2400	3600
Shandong	240	360	420	900	1800	3000	4500
Henan	180	300	420	720	1500	3000	4500
Hebei	120	300	480	840	1800	3000	4500
Hunan	120	300	360	720	1920	3120	4800
Hubei	240	360	420	720	1800	3000	4500
Shanxi	180	300	360	720	1800	3000	4500
Liaoning	300	420	480	900	1800	3000	4500
Jilin	240	420	480	900	1800	3000	4500
Heilongjiang	240	420	480	900	1800	3000	4500
Anhui	180	300	360	660	1200	2700	3900
Fujian	180	300	360	720	1500	2640	3900
Jiangxi	300	420	480	900	1800	3000	4500
Guangxi	60	360	420	780	1800	3000	4500
Sichuan	180	300	360	720	1800	3000	4500
Guizhou	180	300	360	660	1200	2400	3600
Yunnan	60	300	390	780	1800	3000	4500
Shanxi	180	300	480	720	1800	3000	4500
Gansu	240	360	360	360	360	360	360
Ningxia	120	300	360	660	1800	3000	4500
Hainan	60	300	360	720	1500	2700	4200
Inner Mongol	300	360	420	900	1800	3000	4500
Qinghai	60	300	360	660	1500	2700	4200
Xinjiang	180	360	420	720	1800	3000	4500
Xizang	60	300	360	660	1200	2400	3600
Average	179	338	407	747	1606	2816	4227

D represents the engine displacement.

), while BEVs are exempt from the $VUT$, thus resulting in ${VUT}_{BEV}=0$.

According to the national regulation, the license fee for each vehicle is CNY 125, which includes CNY 100 for license plate production, CNY 15 for driver’s license production, and CNY 10 for registration certificate production. Different government subsidies are provided based on the electric driving range. Vehicles with a driving range below 300 km are ineligible for any subsidy. Vehicles with a range between 300 and 400 km receive a subsidy of CNY 9100 per vehicle, while those with a range beyond 400 km receive a subsidy of CNY 12,600 per vehicle.

2.

Infrastructure cost

Both ICEVs and BEVs require specific infrastructure for fueling or charging. China had around 113,000 gas stations in the year 2022, with a minimal increase of merely 12,000 stations between 2018 and 2022. Additionally, the equipment costs for gas stations are significantly lower compared to the costs of charging stations. Therefore, we assume that gas stations have a marginal cost of zero, excluding the infrastructure costs of ICEVs. The EV-pile ratio is considerably lower than that of ICEVs to the gas station, thus the costs of charging points need to be taken into account while assessing BEVs.

A charging point is defined as a terminal apparatus that supplies electrical energy to electric vehicles. By means of plug-in connectors engaging the vehicle’s charging inlet, it enables power transfer and regulates charging processes, thereby constituting the fundamental operational unit of electric vehicle charging infrastructure. The number of charging points varies based on factors such as charging station type, capacity, and location. Therefore, we define the unit of charging infrastructure as a charging point. charging points can be categorized into two types based on the charging technology: alternating current (AC) piles and direct current (DC) piles. Depending on the usage scenario, charging points can be further classified as private and public charging points. There is no strict one-to-one correspondence between vehicles and charging points. Hence, Equation (3) introduces the ratio of charging points to vehicles as ${\mathsf{\omega }}_{\mathrm{i}}$. The infrastructure cost of BEVs is determined by the weighted average of the quantities of private charging points, public AC charging points, and public DC charging points, along with the infrastructure construction cost ($C_{ec}$), infrastructure operating cost ($C_{eo}$), and infrastructure maintenance cost ($C_{em}$). The detailed definition of charging infrastructure cost is as follows.

$$IC=\sum {\omega }_i\times \left(C_{ec,i}+C_{eo,i}+C_{em,i}\right)$$

(3)

where ${\omega }_i$ is the number of different types of charging points required per vehicle, and $i=\mathrm{1,2},3$ refers to private charging points, public AC charging points, and public DC charging points, respectively. According to data released by the China Electric Vehicle Charging Infrastructure Alliance, there were 5.21 million charging points in operation by the end of 2022, with 3.413 million being private charging points and 1.797 million being public charging points. Among the public charging points, there were 0.761 million DC charging points and 1.036 million AC charging points. The total number of BEVs in China was 10.45 million. By the end of 2024, the total number of electric vehicle charging and battery swapping infrastructure in China had reached 12.818 million units. Specifically, the number of private charging points is 8.201 million, and the number of public charging points has reached 4.617 million, including 2.165 million DC charging points and 2.452 million AC charging points [39]. By the end of 2024, the total number of battery electric vehicles in China had exceeded 22.09 million. The ratio of battery electric vehicles to supporting infrastructure in China was approximately 1.72:1, compared with 2.01:1 in 2022 [40].

3.

Infrastructure construction cost ($C_{ec}$)

The infrastructure construction cost ($C_{ec}$) involves purchase cost ($C_{pc})$, installation cost ($C_{ic}$), and land rental cost ($C_{lr}$). The construction cost of charging infrastructure is as follows.

$$C_{ec}=C_{pc}+C_{ic}+C_{lr}$$

(4)

$$C_{lr}=L_p\times L_s$$

(5)

where $L_p$ is the unit area price, and $L_s$ is the area of charging points.

DC charging points are equipped with DC chargers, enabling the direct charging of electric vehicle power batteries. On the other hand, AC charging points require the use of onboard chargers for charging, resulting in significant price differences between the two types of charging points. As detailed in

Table 5. Classification and parameters of charging points.

	Private Charging Points	Public AC Charging Points	Public DC Charging Points
Power of charging points/kW	7	7	60
Charging Time/Hours	5–8	5–8	0.3–2.5
Purchase Cost/CNY	4000	5500	53,300
Installation Cost/CNY	1000	3000	21,000
Land Rent/CNY	0	68,425	68,425
Vehicle-to-Pile/ωi	3.0618	13.7319	10.0869
Operational Time/Year	15	15	15

Data source: Information on Major Equipment and Material Prices for Southern Power Grid Company’s Grid Projects in the Second Quarter of 2022 [41].

, the purchase cost of a 7 kW AC charging point is CNY 5500, whereas the cost of a 60 kW DC charging point is CNY 53,300.

The land rental cost associated with the installation of charging points and provision of parking spaces for vehicle charging is an important consideration. Generally, the land area occupied by a single charging point is not significant. In this study, we consider a standard parking space size as a reference, which is $L_s=12.5 {\mathrm{m}}^2$. To facilitate further research, we adopt the average national land transaction price in 2022 from the “China Land Market Transaction Information of 300 Cities”, published by the Chinese Academy of Sciences, as a benchmark for the unit area price of charging points, denoted as $L_p=$ 5440 $\mathrm{C}\mathrm{N}\mathrm{Y}/{\mathrm{m}}^2$.

4.

Infrastructure operation cost $\left(C_{eo}\right)$

The infrastructure operation cost $\left(C_{eo}\right)$ primarily depends on the cost of providing energy services through charging points. The operation cost of charging infrastructure is as follows.

$$C_{eo}=\sum _{k=1}^n{\displaystyle \frac{Q_{ec}\ast p_e}{(1+r{)}^{k-1}\ast (1-\sigma )}}$$

(6)

$$Q_{ec}=VKT\ast q$$

(7)

where ${\mathrm{Q}}_{\mathrm{e}\mathrm{c}}$ is the charge capacity (kWh), ${\mathrm{p}}_{\mathrm{e}}$ is the purchase price per unit of electricity for the operator (CNY). $\mathsf{\sigma }$ is the charging efficiency, indicating the loss of electrical energy in the form of heat during the charging process. ${\mathrm{Q}}_{\mathrm{e}\mathrm{c}}$ is derived from the multiplication of the electric vehicle’s driving distance ($\mathrm{V}\mathrm{K}\mathrm{T}$) and the average energy consumption per kilometer ($\mathrm{q}$). For this study, we assume a unit electricity price of ${\mathrm{p}}_{\mathrm{e}}=0.672 \mathrm{C}\mathrm{N}\mathrm{Y}/\mathrm{k}\mathrm{W}\mathrm{h}$, a charging efficiency of $\mathsf{\sigma }=8\%$ [42], and an energy consumption of $\mathrm{q}=0.2 \mathrm{k}\mathrm{W}\mathrm{h}/\mathrm{k}\mathrm{m}$ [43].

5.

Infrastructure maintenance cost $\left(C_{em}\right)$

The infrastructure maintenance cost $\left(C_{em}\right)$ refers to the cost of repairing and replacing components during the infrastructure operation. The maintenance cost for infrastructure is typically estimated at 2% of the infrastructure construction cost [42]. Therefore, the maintenance cost of charging infrastructure can be expressed as follows.

$$C_{cm}=\sum _{k=1}^n{\displaystyle \frac{2\%\ast C_{ec}}{(1+r{)}^{k-1}}}$$

(8)

6.

Energy cost

The gasoline and electricity costs are the respective energy costs for ICEVs and BEVs. We consider the consumers’ charging habits to reflect actual electricity costs. Battery electric vehicle owners have the option to choose different charging methods based on their preferences, such as private charging, workplace charging, or public charging. Each charging method has its specific charging price, and therefore, consumer choices directly affect the overall electricity costs. We assume that gasoline and electricity price are certain over the lifetime of the vehicle.

The energy cost for ICEVs can be calculated as

$$C_{e,gv}=VKT\ast p_{gv}\ast Q_{gv}$$

(9)

where $p_{gv}$ is the gasoline price (CNY/L), and ${\mathrm{Q}}_{\mathrm{g}\mathrm{v}}$ is the gasoline consumption per kilometer (L/km). According to the data released by different provinces and cities in China for the average price of 92# gasoline in 2022, we assume that $p_{gv}$ is 7.11 CNY/L, and $Q_{gv}$ is $0.085 \mathrm{L}$/km.

In the case of BEVs, the electricity cost can be expressed as the weighted average of charging costs with and without private charging points [44]. The electricity cost can be written as

$$C_E={RC}_{E,w}+\left(1-R\right)C_{E,o}$$

(10)

$$C_{E,w}=VKT·\left({\xi }_{w,private}·P_{e,private}+{\xi }_{w,public}·P_{e,public}+{\xi }_{w,business}·P_{e,business}\right)$$

(11)

$$C_{E,o}=VKT·\left({\xi }_{w,public}·P_{e,public}+{\xi }_{w,business}·P_{e,business}\right)$$

(12)

where $C_{E,w}$ is the electricity cost for BEVs with private charging points, and $C_{E,o}$ is the electricity cost without private charging points. $R$ is the availability of residential parking spaces, which is primarily determined by the level of urbanization. We obtained fixed values for $R$ in each city for the year 2022 [45]. The electricity parameters for three first-tier cities (Beijing, Shanghai, Guangdong) are presented in

Table 6. Electricity parameters in China’s cities.

City	R (%)	${\mathit{p}}_{\mathit{e},\mathit{p}\mathit{u}\mathit{b}\mathit{l}\mathit{i}\mathit{c}}+{\mathit{p}}_{\mathit{s},\mathit{p}\mathit{u}\mathit{b}\mathit{l}\mathit{i}\mathit{c}}$ (CNY/kWh)	${\mathit{P}}_{\mathit{e},\mathit{b}\mathit{u}\mathit{s}\mathit{i}\mathit{n}\mathit{e}\mathit{s}\mathit{s}}$ (CNY/kWh)	${\mathit{P}}_{\mathit{e},\mathit{p}\mathit{r}\mathit{i}\mathit{v}\mathit{a}\mathit{t}\mathit{e}}$ (CNY/kWh)
Beijing	70.7	1.6	0.8	0.6
Shanghai	57.9	2.5	0.9	0.4
Guangdong	91.5	1.8	0.8	0.7
Average	60.0	2.0	0.8	0.6

R represents the availability of residential parking spaces.

. $P_{e,private}$, $P_{e,public}$ and $P_{e,business}$ represent the electricity prices of using private charging points, public charging points and business charging points, respectively. Considering that private charging usually occurs after work hours, the price for private charging points is determined based on the peak electricity price used by residents [46]. As for public charging, the charging price consists of the base electricity fee and the charging service fee, which can be calculated using the following formulation.

$$P_{e,public}=p_{e,public}+p_{s,public}$$

(13)

where $p_{e,public}$ is the base electricity fee, $p_{s,public}$ is the charging service fee.

${\xi }_{w,private}$, ${\xi }_{w,public}$ and ${\xi }_{w,business}$ represent the probabilities of using private charging points, public charging points, and business charging points, respectively (

Table 7. Based on the hypothesis of charging location probabilities.

Locations	Drivers with Private Charging Points ${\mathit{\xi }}_{\mathit{w}}$	Drivers Without Private Charging Points ${\mathit{\xi }}_{\mathit{o}}$
Home parking	60%	0%
Public charging	20%	60%
Business	20%	40%

).

7.

Emission cost

Given the potential for reductions in GHG emissions may be offset by harmful increases in particular matters emissions, emission cost can be considered as an important dimension in cost-effectiveness analysis. Two types of emission costs are considered: air quality emission costs and greenhouse gas emission costs. Air quality emission costs encompass the impacts on human health and ecosystems. We mainly focus on the emission costs of five major pollutants, including $\mathrm{C}\mathrm{O},{\mathrm{N}\mathrm{O}}_{\mathrm{x}},\mathrm{V}\mathrm{O}\mathrm{C},{\mathrm{S}\mathrm{O}}_{\mathrm{x}},{\mathrm{P}\mathrm{M}}_{2.5},$ and ${\mathrm{P}\mathrm{M}}_{10}$. On the other hand, greenhouse gas emission costs relate to the emissions that contribute to global warming. We mainly examine three major pollutants, including ${\mathrm{C}\mathrm{O}}_2, {\mathrm{C}\mathrm{H}}_4, \mathrm{a}\mathrm{n}\mathrm{d}$ ${\mathrm{N}}_2\mathrm{O}$. The emission costs are calculated based on the emissions of various air pollutants and their unit costs and can be expressed as

$$C_{EMC}=C_{AQ}+C_{GHG}$$

(14)

$$C_{AQ}=\sum _{a\in AQ}{P}_a\ast e_a\ast VKT$$

(15)

$$C_{GHG}=\sum _{bϵGHG}{P}_b\ast e_b\ast VKT$$

(16)

where $C_{AQ}$ and $C_{GHG}$ represent the emission costs of air quality and greenhouse gases, respectively (CNY). $\mathrm{P}$ is the emission quantity of the related pollutants (g/km). Pollutant equivalence refers to the relative value of a specific pollutant in terms of its negative impact on the environment and health and is converted into a comparable and unified unit through scientific assessment. Specifically, the calculation of pollutant equivalents can refer to the Environmental Protection Tax Law and relevant technical specifications. For instance, the emissions of major pollutants such as sulfur dioxide (SO₂) and nitrogen oxides (NOx) can be converted into standard equivalent values based on their hazard coefficients. The calculation of external cost (e) is achieved by multiplying the number of pollutant equivalents by the environmental tax rate per unit equivalent (CNY/g). This tax rate comprehensively considers economic parameters such as pollution control costs and ecological damage values. $\mathrm{A}\mathrm{Q}$ is the set of pollutants that impact air quality, and $\mathrm{G}\mathrm{H}\mathrm{G}$ is the set of pollutants that affect greenhouse gases. The environmental tax amount for each province and the pollutant equivalent unit can be seen in

Table 8. The environmental tax amount for each province and the pollutant equivalent unit.

	$\mathbf{C}\mathbf{O}$	${\mathbf{N}\mathbf{O}}_{\mathbf{x}}$	$\mathbf{V}\mathbf{O}\mathbf{C}$	${\mathbf{S}\mathbf{O}}_{\mathbf{x}}$	${\mathbf{P}\mathbf{M}}_{2.5}$	${\mathbf{P}\mathbf{M}}_{10}$	${\mathbf{C}\mathbf{O}}_2$	${\mathbf{C}\mathbf{H}}_4$	${\mathbf{N}}_2\mathbf{O}$
Beijing	12	12	12	12	12	12	12	12	12
Tianjin	10	10	10	10	10	10	10	10	10
Hebei	6	6	6	6	6	6	6	6	6
Shandong	1.2	6	1.2	6	1.2	1.2	1.2	1.2	1.2
Shanxi	1.8	1.8	1.8	1.8	1.8	1.8	1.8	1.8	1.8
Liaoning	1.2	1.2	1.2	1.2	1.2	1.2	1.2	1.2	1.2
Heilongjiang	1.2	1.2	1.2	1.2	1.2	1.2	1.2	1.2	1.2
Jilin	1.2	1.2	1.2	1.2	1.2	1.2	1.2	1.2	1.2
Inner Mongol	2.4	2.4	2.4	2.4	2.4	2.4	2.4	2.4	2.4
Xinjiang	1.2	1.2	1.2	1.2	1.2	1.2	1.2	1.2	1.2
Ningxia	1.2	1.2	1.2	1.2	1.2	1.2	1.2	1.2	1.2
Xizang	1.2	1.2	1.2	1.2	1.2	1.2	1.2	1.2	1.2
Jiangxi	1.2	1.2	1.2	1.2	1.2	1.2	1.2	1.2	1.2
Jiangsu	8.4	8.4	8.4	8.4	8.4	8.4	8.4	8.4	8.4
Shanghai	1.2	8.55	1.2	7.6	1.2	1.2	1.2	1.2	1.2
Henan	4.8	4.8	4.8	4.8	4.8	4.8	4.8	4.8	4.8
Gansu	1.2	1.2	1.2	1.2	1.2	1.2	1.2	1.2	1.2
Guangxi	1.8	1.8	1.8	1.8	1.8	1.8	1.8	1.8	1.8
Guangdong	1.8	1.8	1.8	1.8	1.8	1.8	1.8	1.8	1.8
Chongqing	3.5	3.5	3.5	3.5	3.5	3.5	3.5	3.5	3.5
Hainan	2.4	2.4	2.4	2.4	2.4	2.4	2.4	2.4	2.4
Anhui	0.6	1.2	1.2	1.2	0.6	0.6	0.6	0.6	0.6
Fujian	1.2	1.2	1.2	1.2	1.2	1.2	1.2	1.2	1.2
Yunnan	2.8	2.8	2.8	2.8	2.8	2.8	2.8	2.8	2.8
Sichuan	3.9	3.9	3.9	3.9	3.9	3.9	3.9	3.9	3.9
Hunan	2.4	2.4	2.4	2.4	2.4	2.4	2.4	2.4	2.4
Hubei	2.8	2.8	2.8	2.8	2.8	2.8	2.8	2.8	2.8
Zhejiang	1.2	1.2	1.2	1.2	1.2	1.2	1.2	1.2	1.2
Qinghai	1.2	1.2	1.2	1.2	1.2	1.2	1.2	1.2	1.2
Shanxi	1.2	1.2	1.2	1.2	1.2	1.2	1.2	1.2	1.2
Guizhou	1.2	1.2	1.2	1.2	1.2	1.2	1.2	1.2	1.2
Average	2.75	3.17	2.77	3.14	2.75	2.75	2.75	2.75	2.75
Pollutant equivalent unit	16.7	0.95	0.95	0.95	4	4	20	0.67	0.95

The unit of environmental tax amount in Table 8 is yuan per pollutant equivalent (CNY/g), which is used to measure the standard for levying taxes and fees on the discharge of specific pollutants, reflecting the amount of environmental tax to be paid for each pollutant equivalent discharged.

.

8.

Maintenance cost

Battery maintenance is the major maintenance cost for BEVs due to the battery capacity gradually decreases over time. For ICEVs, the engine and fuel system are commonly the focus of maintenance and replacement efforts. We assume that the maintenance cost remains constant during the ownership period, and maintenance is performed every 5000 km. The average maintenance cost of ICEVs is CNY 650 per 5000 km, while it is CNY 400 per 5000 km for BEVs [47].

9.

Scrap value

After a vehicle reaches a certain lifespan or mileage, its reusable components are dismantled and recycled. When the battery capacity drops below 80% of its initial capacity, it enters the recycling stage [48]. However, it is important to note that the current absence of a standardized recycling mechanism in China may affect the accuracy of estimating the recycling cost of vehicles and batteries. The vehicle scrap value is given by

$$C_R={\left(1-w\right)}^n\ast MSRP+C_B$$

(17)

where $\mathrm{w}$ is the annual depreciation rate, which is assumed to be constant. In this case, $w$ is set to a value of 20% [49]. $C_B$ is the battery cost, which is CNY 8000 [50]. Additionally, we assume that the depreciation rates for both BEVs and ICEVs are the same for the purpose of simplifying the model.

2.3.3. Cost-Effectiveness

The reduction of GHG emissions over the process of a product’s life cycle often results in an increase in life cycle cost. The cost-effectiveness is defined to be the cost required to switch from ICEVs to BEVs to reduce a unit of GHG emissions. Then the cost-effectiveness is calculated by

$$CE={\displaystyle \frac{C_{BEV}-C_{ICEV}}{{GHG}_{ICEV}-{GHG}_{BEV}}}$$

(18)

where $C_{BEV}$ is the life cycle cost of BEVs (CNY), $C_{ICEV}$ is the life cycle cost of ICEVs (CNY), ${GHG}_{ICEV}$ is the GHG emissions of ICEVs (kg), and ${GHG}_{BEV}$ is the GHG emissions of BEVs (kg). Cost difference refers to the incremental cost incurred in the process of converting from ICEVs to BEVs to reduce GHG emissions. It represents the comprehensive financial impact of the shift from ICEVs to BEVs in the pursuit of GHG reduction. The cost-effectiveness value can be either positive or negative, depending on the degree of cost difference and GHG reduction. When the cost difference is positive, and the GHG emissions of ICEVs are higher than those of BEVs, it indicates a reduction in GHG emissions at a higher cost. When this occurs, a lower CE value is preferred. On the other hand, if the cost difference is negative, and the GHG emissions of ICEVs are higher than those of BEVs, it indicates achieving low carbon emissions at a lower cost. In this scenario, a larger CE value is preferred, representing the most ideal situation in which battery electric passenger cars serve as an alternative for gasoline vehicles.

2.4. Scenario Analysis

We propose three low-carbon measures including electricity decarbonization (ED), energy efficiency improvement (EEI), and vehicle lightweight (LW). Based on these three measures and three different levels of technological development, nine scenarios are designed in this study, as shown in

Table 9. Characteristics of different scenarios.

	Low Speed			Medium Speed			High Speed
	ED	EEI	LW	ED	EEI	LW	ED	EEI	LW
BAU
L1	+
L2	+	+
L3	+	+	+
M1				+
M2				+	+
M3				+	+	+
H1							+
H2							+	+
H3							+	+	+

. These scenarios, named L1, L2, and L3, M1, M2, and M3, H1, H2, and H3, provide a framework for predicting the trends of carbon reduction potential and cost-effectiveness of electric passenger cars for the years 2025, 2030, and 2035. As a reference, we set a baseline scenario called BAU, assuming that the future vehicle technology level remains the same as the year 2022. The L1 scenario includes electricity decarbonization measures in addition to the BAU scenario under a low level of technological development, while the L2 scenario further improves the fuel economy based on the L1 scenario, and the L3 scenario combines all three measures. To ensure the representativeness of the results, the analysis selects the A-class sedan, B-class sedan, A-class SUV, and B-class SUV for comparison in these different scenarios.

By considering different scenarios and their effects on carbon emissions and cost-effectiveness, policymakers and OEMs can gain insights into the potential benefits and challenges. This analysis will help in managing uncertainties, optimizing resource allocation, and guiding decision-making in the electric vehicle market.

2.4.1. Electricity Decarbonization—ED

In recent years, the Chinese government has implemented a series of measures to promote low-carbon energy transformation. Currently, coal-fired power generation holds a dominant position within China’s electricity structure. Hence, the expected gradual increase in the share of clean energy sources in the future, makes predictions on electricity decarbonization necessary. Based on research conducted by the National Development and Reform Commission of China and the relevant literature ([51,52,53]) we have set future scenarios for the electricity structure in 2025, 2030, and 2035; see

Table 10. Parameter settings for ED.

Electricity Mix (%)	Baseline	Low Speed			Medium Speed			High Speed
	Year 2022	Year 2025	Year 2030	Year 2035	Year 2025	Year 2030	Year 2035	Year 2025	Year 2030	Year 2035
Oil electricity	0.0%	0.0%	0.0%	0.0%	0.0%	0.0%	0.0%	0.0%	0.0%	0.0%
Gas electricity	0.0%	0.0%	0.0%	0.0%	12.00%	14.50%	15.60%	10.0%	13.4%	16.8%
Coal electricity	58.40%,	55.0%	49.0%	40.0%	46.90%	38.60%	28.00%	41.7%	32.2%	22.7%
Nuclear electricity	4.72%	7.0%	8.0%	9.0%	2.60%	4.10%	5.60%	11.8%	15.2%	18.6%
Biomass	0.0%	0.0%	0.0%	0.0%	0.0%	0.0%	0.0%	4.0%	5.2%	6.4%
Others	36.88%	38.0%	43.0%	51.0%	38.50%	42.80%	50.80%	32.5%	34.0%	35.5%

.

2.4.2. Energy Efficiency Improvement—EEI

Carbon emissions are directly influenced by fuel economy or 100 km electricity consumption, which refers to the amount of fuel or electricity consumed by a vehicle to cover a certain distance. Vehicles with higher energy efficiency consume less fuel or electricity while traveling the same distance, resulting in reduced carbon emissions and lower energy costs. According to the plan of “Technology Roadmap for Energy Saving and New Energy Vehicles 2.0”, we have made predictions about the energy efficiency of BEVs based on a medium development rate. Moreover, we assume that energy efficiency will experience a gradual increase or decrease at an average rate of 0.5 kWh/100 km every 5 years under both low-speed and high-speed conditions, respectively, compared to the medium development rate. The energy efficiency scenario parameter settings as shown in

Table 11. Parameter settings for EEI and LW.

Electric Vehicle		Low Speed			Medium Speed			High Speed
		Year 2025	Year 2030	Year 2035	Year 2025	Year 2030	Year 2035	Year 2025	Year 2030	Year 2035
Energy efficiency (kWh/100 km)	A	11.5	11	10.5	11	10.5	10	10	9.5	9
	B	13.5	13	12.5	13	12.5	12	12.5	12	11.5
Vehicle lightweight degree (%)		10	20	30	15	25	35	20	30	40

.

2.4.3. Vehicle Lightweight—LW

Reducing the curb weight can improve vehicles’ fuel economy by decreasing the demand for fuel, thus resulting in decreased fuel consumption and carbon emissions [54]. Moreover, lightweight has the additional advantage of reducing overall vehicle weight, which leads to improved battery efficiency and a longer driving range [55]. However, a lighter curb weight also costs much because of substituted materials. According to the plan of “Technology Roadmap for Energy Saving and New Energy Vehicles 2.0”, we have generated projections about the vehicle lightweight degree based on a medium development rate. Furthermore, we posit that the lightweight degree would decrease or increase by 5% every 5 years under low-speed and high-speed conditions, respectively, in comparison to the medium development rate. The vehicle lightweight degree scenario parameter settings are shown in Table 11.

3. Results

3.1. Life Cycle Emission Results

Figure 2 illustrates the life cycle carbon emissions of the selected vehicles. These emissions vary significantly based on the vehicles’ segments, classes and driving ranges are different. BEVs emit 12.908–37.422 t${\mathrm{C}\mathrm{O}}_2$, eq/vehicle, while ICEVs emit 25.032–48.779 t${\mathrm{C}\mathrm{O}}_2$, eq/vehicle. The carbon emissions of BEVs are 23.28–48.43% lower than those of ICEVs. The results show that carbon emissions from electric passenger cars are on average reduced by about 31.85% compared to ICEVs, which highlights the significant advantage of electric passenger cars in lowering carbon emissions.

BEV200: The driving range is between 200 km (inclusive) and 300 km (exclusive). BEV300: The driving range is between 300 km (inclusive) and 400 km (exclusive). BEV400: The driving range is between 400 km (inclusive) and 500 km (exclusive). BEV500: The driving range is over 500 km (inclusive).

Carbon reduction rates vary significantly across different vehicle segments. SUVs have lower reduction rates than sedans in general. The main reason is the increase in curb weight raises carbon emissions directly. In terms of sedans, A00-class sedans stand out with the lowest carbon emissions and the highest carbon reduction rate, reaching up to 50.97%. A key factor behind this achievement is the lightweight and compact nature of the A00-class sedan, which reduces air resistance and friction, thereby leading to lower carbon emissions.

This study further reveals that a longer driving range does not correspond to higher carbon emissions. The main reason is that the electric driving range is primarily determined by battery capacity. A longer driving range requires a larger battery capacity, which leads to the installation of more batteries. As a result, this increases curb weight and power consumption, subsequently raising carbon emissions. In summary, it can be concluded that driving range is not a direct factor influencing carbon emissions.

Based on the carbon emissions during the life cycle of the selected vehicles, we further selected A00-class sedans, A0-class sedans, A-class sedans, A00-class SUVs, A0-class SUVs, and A-class SUVs for sensitivity analysis related to the power structure. The results are shown in

Table 12. Sensitivity analysis of the impact of power structure on carbon emissions.

	Segment	Class	CED (t)	ACR-P	OCR-P	ICE (t)	ACR-B	OCR-B
BEV	Sedan	A00	9	61.61%	48.73%	12.91	52.57%	31.04%
		A00	9.9			14.06
		A00	11.85			16.64
		A0	11.1	51.41%		15.60	43.88%
		A0	12.6			17.59
		A0	15			21.07
		A	12.45	53.26%		17.51	35.59%
		A	13.35			19.00
		A	13.95			19.48
	SUV	A0	15.3	35.42%		15.99	33.79%
		A0	16.8			18.68
		A0	18.15			19.84
		A0	19.5			25.53
		A	17.4	41.95%		28.76	−10.66%
		A	19.05			26.58
		A	18.15			27.76
ICEV	Sedan	A00	26.7			30.65
		A0	26.55			32.23
		A	28.35			28.98
	SUV	A0	27			30.22
		A	31.35			25.03

CED: carbon emissions after decarbonization. ICE: initial carbon emissions (tons). ACR-P: abbreviation of average carbon reduction rate-post-decarbonization, meaning the average carbon emission reduction rate of BEVs compared to conventional ICEVs of the same type and class after electricity decarbonization. OCR-P: abbreviation of overall carbon reduction rate-post-decarbonization, meaning the overall average carbon emission reduction rate of battery electric vehicles compared to conventional internal combustion engine vehicles of the same type and class after electricity decarbonization. ACR-B: abbreviation of average carbon reduction rate-post-baseline, meaning the average carbon emission reduction rate of battery electric vehicles compared to conventional internal combustion engine vehicles of the same type and class at the initial stage. OCR-B: abbreviation of the overall average carbon emission reduction rate of battery electric vehicles compared to internal combustion engine vehicles of the same type and class at the initial stage.

. In the case of a clean power structure in the future, we assume that China’s power structure still relies on coal, accounting for 42% [56], nuclear power for 8% [57], and other energies for 50%. The rest of the conditions remain unchanged.

The analysis results are shown in Table 12. The emissions of battery electric vehicles are 9–19.5 t${\mathrm{C}\mathrm{O}}_2$, eq/vehicle, while those of internal combustion engine vehicles are 26.55–31.35 t${\mathrm{C}\mathrm{O}}_2$, eq/vehicle. The carbon emissions of battery electric vehicles are 35.42–61.61% lower than those of internal combustion engine vehicles. The results show that the carbon emissions of the selected grade of electric passenger vehicles have been reduced by an average of approximately 48.73% compared to internal combustion engine vehicles, and there is a significant increase compared to the emission reduction rate of 31.04% of the selected grade of electric passenger vehicles in the original electric structure. This is because a large amount of electrical energy is consumed in the production process of electric vehicles, especially in the production of batteries. The decarbonization of the power structure can reduce carbon emissions in the battery production process, thereby further reducing carbon emissions throughout the entire life cycle of electric vehicles. At the same time, this confirms the conclusion that electric passenger vehicles have significant advantages in reducing carbon emissions. After the decarbonization of the power structure, the emission reduction rate of SUVs is still lower than that of sedans because the decarbonization of the power structure does not cause changes in curb weight and thus cannot have a significant impact on the comparison of their emission reduction capabilities. The carbon reduction rate of A00-class sedans remains the most considerable, reaching 61.61%. It can be seen that the lightweighting and compactness of vehicle bodies are important ways to reduce carbon emissions.

The results of the sensitivity analysis further confirm that the driving range is not a direct factor affecting carbon emissions. For instance, the BEV400 among A-class SUVs generates more carbon emissions compared to the BEV300 and BEV500. It is worth noting that the emission reduction rate of A-class SUVs has increased from −10.65% to 41.95% after the decarbonization of the power structure, indicating that this type of vehicle is highly sensitive to changes in the power structure.

3.2. Life Cycle Cost Results

The life cycle costs of the selected vehicles are shown in Figure 3. The results vary based on powertrain systems, vehicle segments, classes, and driving ranges. The results indicate that BEVs and ICEVs are not economically comparable in the Chinese automotive market. The life cycle costs of ICEVs range from CNY 182,648 to CNY 347,214, while the life cycle costs of BEVs range from CNY 189,954 to CNY 583,320, approximately 1.04 to 1.68 times higher than ICEVs. Due to lower electricity prices, BEVs have an advantage over ICEVs in terms of energy costs, but this advantage is not enough to compensate for the MSRP difference, as the MSRP of a BEV is nearly double that of an ICEV.

The charging infrastructure costs account for around 18.8–57.8% of the overall life cycle costs, with an average allocation of CNY 109,870.7 per vehicle. As the charging infrastructures advance, it is expected that the infrastructure costs will decline in the future. National subsidies do not significantly lower the life cycle costs of BEVs compared to ICEVs, accounting for only 2% to 5% of the overall life cycle costs. This shows that the current national subsidies for BEVs are quite low in terms of economic feasibility. The effect of purchase tax and vehicle tax exemptions can also be ignored.

SUVs have larger bodies and higher technical requirements for chassis and engines, resulting in higher initial purchase costs compared to sedans. Although vehicles with longer driving ranges are available for higher government subsidies, the increase in battery capacity that comes with longer driving ranges leads to higher battery costs. Thus, it can be observed that vehicles with longer driving ranges have more expensive life cycle costs.

This research further explores in depth the scenario of reducing the construction cost of charging points in the short term (1–3 years). In the analysis of the full life cycle cost of electric vehicles, the cost of charging infrastructure is a key factor. In the sensitivity analysis, the following assumptions can be made: (1) As the market size of charging points gradually expands, the procurement cost of charging points decreases. (2) Through continuous installation practice, the enterprise has accumulated rich experience, optimized the installation process, and thus reduced the installation cost. (3) The extent of cost reduction is divided into two scenarios: low-speed and high-speed. The number of charging points and battery electric vehicles is based on the data of China in 2024. (4) Other changes in infrastructure costs are not taken into account in the short term.

The analysis results show that in the low-speed scenario, both the procurement cost and the installation cost are reduced by 5%. When only the impact of short-term scale effect is considered, the average cost of charging infrastructure per vehicle is CNY 126,920.14, an increase of 15.52% compared with the base period. When considering the synergy of scale effect and experience accumulation simultaneously, the cost of charging infrastructure accounts for approximately 21.12–61.29% of the total life cycle cost of battery electric vehicles, with an average of CNY 126,682.02 per vehicle, an increase of 15.39%. The gap between the two is relatively small, indicating that the positive feedback loop of the learning effect and scale expansion is not significant in the short term. The main reason lies in the small time span of the research and the installation cost being only 25–55% of the procurement cost.

In the high-speed scenario, the procurement cost is reduced by 20%, while the installation cost is reduced by 15%. When the synergy of the two is taken into account simultaneously, the cost of charging infrastructure accounts for approximately 20.94–61.03% of the total life cycle cost of battery electric vehicles, with an average of CNY 125,407.88 per vehicle, an increase of 14.14%. The growth rate was slightly lower compared to the low-speed situation, indicating that under the set cost reduction range, the synergy of scale effect and experience accumulation makes it difficult to reduce the cost of charging infrastructure in the short term. The main reason is that the number of battery electric vehicles and their charging infrastructure has increased rapidly in two years, but the growth rate of the number of charging points is obviously greater than that of battery electric vehicles. This has led to an increase in the average cost ultimately allocated to each vehicle.

3.3. Cost-Effectiveness Results

The cost-effectiveness of GHG emissions reduction and cost differences for different vehicles are presented in Figure 4. From the standpoint of the vehicle segments, overall cost-effectiveness for sedans is lower than that for SUVs, because sedans are cheaper in MSRP and emit fewer GHGs than SUVs.

For sedans, A-class and B-class sedans both have lower cost-effectiveness than ICEVs across different driving ranges, indicating that they can be used as alternatives to ICEVs. But the cost-effectiveness of BEV400 for A00-class sedans, BEV500 for A0-class sedans, and BEV400 for C-class sedans are worse, due to these vehicles having higher power consumption and initial purchase costs. From Figure 4, it is further evident that BEVs with shorter driving ranges tend to be more cost-effective. The cost-effectiveness of BEVs decreases as driving range increases for two reasons: First, BEVs with shorter driving ranges often have lower LCC, and shorter driving ranges imply smaller battery capacity and lower battery cost, which are major components of electric vehicle costs. Second, BEVs with shorter driving ranges commonly use lower-power charging infrastructures, such as home charging stations, which can reduce the construction and maintenance costs of charging infrastructures.

For SUVs, the BEV300 has poorer cost-effectiveness, the BEV400 is the most cost-effective choice, and the BEV500 is barely cost-effective. This phenomenon is mostly caused by the fact that SUVs are designed for long-distance travel, where shorter driving ranges may limit their convenience. Low-driving-range vehicles are gradually being phased out of the real-world automotive market. Additionally, the emission reduction rate of SUVs gradually decreases as the driving range increases, while increasing battery capacity results in higher costs. As a result, BEV500 shows poorer cost-effectiveness compared to BEV400. The cost-effectiveness of C-class SUVs is negligible.

3.4. Carbon Reduction Potential Results

To determine the relationship between the life cycle carbon emissions and performance parameters of vehicle models, we conducted a multivariate correlation analysis. We examined the relationship between carbon emissions and battery capacity, curb weight, battery weight, fuel economy, driving mileage, and energy density, as shown in Figure 5. The correlation coefficients show that carbon emissions have a strong association ($p<0.05$) with battery capacity, curb weight, battery weight, fuel economy, and driving mileage, while the correlation with battery energy density is not significant. There is a strong positive correlation between carbon emissions and both fuel economy and curb weight, with correlation coefficients of $0.92$ for fuel economy and $0.85$ for curb weight. These results provide support for the selection of influencing factors in scenario analysis.

To determine the relationship between the life cycle carbon emissions and performance parameters of vehicle models, we conducted a multivariate correlation analysis. In order to ensure the reliability of the statistical results and avoid the risk of Type I errors, we implemented a Bonferroni correction for multiple comparisons [58]. Specifically, given that we were examining the relationships between carbon emissions and six performance parameters (battery capacity, curb weight, battery weight, fuel economy, driving mileage, and energy density), we adjusted the significance level. The original significance level of p < 0.05 was divided by the number of comparisons (21 in this case), resulting in a more stringent adjusted significance level of p < 0.0024 (

Table 13. Carbon emissions–performance parameter correlation Bonferroni correction.

	Battery Capacity	Curb Weight	Battery Weight	Emissions	Fuel Economy	Driving Mileage	Energy Density
Battery capacity		0.000	0.015	0.000	0.000	0.000	0.015
Curb weight	0.000		0.000	0.000	0.000	0.000	0.191
Battery weight	0.000	0.000		0.000	0.000	0.000	0.650
Emissions	0.000	0.000	0.000		0.000	0.044	0.317
Fuel economy	0.000	0.000	0.000	0.000		0.045	0.384
Driving mileage	0.000	0.000	0.000	0.044	0.045		0.003
Energy density	0.015	0.191	0.650	0.317	0.384	0.003

).

After the Bonferroni correction, the correlation coefficients show that carbon emissions have a strong association with battery capacity, curb weight, battery weight, and fuel economy, while the correlation with battery energy density and driving mileage is not significant. These results provide support for the selection of influencing factors in scenario analysis.

Figure 6, Figure 7, Figure 8 and Figure 9 illustrate the decomposition of potential emission reduction contributed by each low-carbon measure, where four vehicle models with better cost-effectiveness are selected. Based on a comparative analysis of three low-carbon emission reduction measures, it is evident that a synergistic approach that combines electricity decarbonization, energy efficiency improvements, and vehicle lightweight could yield greater environmental benefits compared to the implementation of a single emission reduction measure. In the comprehensive adoption of the three low-carbon measures, electricity decarbonization emerges as the most potent in reducing carbon emissions, followed by improvement in energy efficiency and vehicle lightweight.

3.4.1. Electricity Decarbonization Carbon Reduction Potentials

Increasing the share of renewable energy in the electric grid can effectively reduce the carbon footprint of electricity, thereby mitigating carbon emissions from vehicles. Electricity decarbonization measure holds the greatest potential for reducing carbon emissions. By increasing the proportion of clean energy generation to about 60%, as projected in 2025, a carbon reduction potential of 45.3% would be achieved. Take the A-class sedan as an example. Compared to the BAU scenarios, ED further decreases the carbon emissions by 17.1%, 27.1%, and 36.6%, respectively. This means that a significant portion of carbon emissions can be eliminated by transitioning to cleaner sources of electricity.

Figure 10, Figure 11, Figure 12 and Figure 13 show the contribution of the ED measure to the emission reductions across different stages. With the deep decarbonization of electricity, the sedan A-class with an electric range of 400 km exhibits emission reductions of 2486 kg, 4142 kg, and 5803 kg in the WTP stage for the years 2025, 2030, and 2035, respectively, while the vehicle cycle would only contribute to emission reductions of 359 kg, 460 kg, and 838 kg, respectively. Deep decarbonization leads to significant reductions in carbon emissions from electricity generation, resulting in the majority of life-cycle emission reductions occurring in the WTP stage. This is due to the significant clean energy generation in the power grid, which result in very low carbon intensity. Although the vehicle cycle also contributes to emission reductions, its impact is relatively minor.

3.4.2. Energy Efficiency Improvement Carbon Reduction Potentials

EI measure decreases electricity consumption per km. The energy efficiency improvement measure demonstrates significant carbon reduction potentials, ranking second only to electricity decarbonization. It further demonstrates a direct correlation between energy efficiency and the carbon emissions of electric vehicles. Take the A-class sedan with an electric range of 400 km as an example. Compared to the H1 scenarios, improving energy efficiency from the current levels to 10, 9.5, and 9 is estimated to decrease carbon emissions in the H2 scenarios by 6.6%, 7.3%, and 7.8%, respectively. Achieving a carbon reduction of up to 14.9% is possible by improving energy efficiency to 9 kWh/100 km.

Figure 14, Figure 15, Figure 16 and Figure 17 show the contribution of the EEI measure to the emission reductions across different stages. In the EEI scenarios, the carbon reduction potential of BEVs is only observed during the WTP stage. Improved energy efficiency leads to lower electricity demand and decreased reliance on conventional fossil fuels, resulting in significant emission reductions in the future. For example, sedan A-400 exhibits emission reductions of 1554 kg, 2072 kg, and 2590 kg in the WTP stage for the years 2025, 2030, and 2035, respectively, while the EEI has no effect in the vehicle cycle.

3.4.3. Vehicle Lightweight Carbon Reduction Potentials

Increasing the vehicle lightweight degree positively impacts carbon emission reduction, reinforcing the findings presented in Figure 5. Take the A-class sedan with electric range of 400 km as an example. Compared to the H2 scenarios, the cumulative carbon reduction percentages under vehicle lightweight degrees of 20%, 30%, and 40% is 5.5%, 6.3%, and 7.0%, respectively. The utilization of lightweight materials enables a carbon reduction potential of 9.0%. In addition, among the current emission reduction paths, achieving vehicle lightweight is the most accessible, with relatively high potential. Increasing electric driving range and reducing battery costs are important contributions to vehicle lightweight.

Figure 18, Figure 19, Figure 20 and Figure 21 shows the contribution of LW measure to the emission reductions across different stages. The LW measure has a complex impact on emission reductions. In contrast to the ED and EEI measures, the carbon reductions achieved in the WTP stage are partially offset by the increased carbon emissions in the vehicle cycle. For example, with the low technical development level, implementing vehicle lightweight measures would lead to emission reductions of 1896 kg, 1987 kg, and 2072 kg during the WTP. However, there would be an increase in emissions of 1628 kg, 942 kg, and 256 kg during the vehicle cycle for A-class sedans for the years 2025, 2030, and 2035, respectively. Notably, the increase in carbon emissions during the vehicle cycle would improve as technology improves. On one hand, lighter vehicles require less energy to propel, resulting in reduced fuel consumption and carbon emissions. LW therefore reduces carbon emissions and improves fuel economy in the WTP stage. On the other hand, the use of lightweight materials may have a higher environmental impact compared to traditional materials, increasing carbon emissions in the vehicle cycle. Therefore, observing the environmental impacts of vehicle lightweight technologies needs to be considered from a life cycle perspective.

4. Conclusions

A cost-effectiveness analysis is conducted to compare the carbon reduction potentials and economic viability of BEVs and ICEVs in the Chinese market, considering different powertrains, body structures, vehicle classes, and driving ranges. The goal of this study is to identify the most cost-effective vehicle. Additionally, we propose three low-carbon measures, including electricity decarbonization, energy efficiency improvement, and vehicle lightweight, to forecast the change trends of emission mitigation potentials in the years 2025, 2030, and 2035. The main conclusions are as follows:

(1)

From the standpoint of carbon emissions, BEVs demonstrate considerable advantages in energy saving and carbon reduction. Compared to ICEVs, BEVs exhibit an average reduction of about 31.85% in carbon emissions, with A00-class sedans and B-class SUVs showing the highest carbon reduction rates. The primary contributor of GHG emissions for all vehicles is the fuel cycle stage, with emissions roughly twice as high as the vehicle cycle stage of BEVs, and roughly seven times higher than the vehicle cycle stage of ICEVs.

(2)

From the standpoint of life cycle cost, BEVs and ICEVs are not economically comparable in China. The lower energy costs of BEVs are insufficient to offset the MSRP gap. Charging infrastructure costs account for about 18.8–57.8% of the vehicle’s life cycle costs. The effects of national subsidies and tax reduction policies (such as purchase taxes and vehicle taxes) are negligible.

(3)

From the standpoint of cost-effectiveness, overall sedans have lower cost-effectiveness compared to SUVs. Both A-class and B-class sedans demonstrate lower cost-effectiveness than ICEVs across different driving ranges, indicating that A-class and B-class sedans have the potential to serve as alternatives to ICEVs. For SUVs, the BEV400 proves to be the most cost-effective choice.

(4)

In three low-carbon scenarios, electricity decarbonization emerges as a crucial factor in mitigating carbon emissions, followed by improvement in energy efficiency and vehicle lightweight. BEVs could achieve a carbon reduction potential of up to 45.3%, 14.9%, and 9.0% in the ED, EEI, and LW scenarios, respectively. Moreover, adopting a synergistic approach by combining these three measures could yield greater emission mitigation benefits compared to implementing a single emission reduction measure.

Notably, each measure exhibits distinct impact on different stages. The ED measure predominantly influences the WTP stage, with the vehicle cycle being affected subsequently, and EEI measure only affects the WTP stage. On the other hand, the LW measure would result in emission reductions in the WTP stage and emission increases in 111 the vehicle cycle due to the use of lightweight materials.

The research findings have several policy implications. Firstly, it is recommended to enhance government support for NEVs by providing purchase subsidies and tax exemptions and establishing charging infrastructure, to reduce the ownership and usage costs of NEVs, and to facilitate the development of the new energy vehicle market.

Secondly, policymakers should consider eliminating the minimum driving range restrictions and incentivizing manufacturers to produce A-class sedans, B-class sedans, and SUVs with a 400 km driving range. The current government subsidies, which are linked to the electric driving range, may encourage the production of BEVs with larger battery capacities. This can potentially escalate the environmental impact of battery production. Therefore, the Chinese government should consider several factors, including price, performance, environmental friendliness, and energy efficiency when formulating subsidy policies for BEVs.

Finally, it is crucial to accelerate decarbonization in the power sector by increasing the share of clean energy generation and reducing reliance on fossil fuels to lower carbon emissions during electricity production. At the same time, there should be a focus on encouraging automobile manufacturers to adopt advanced energy-saving technologies and lightweight materials, to improve the overall energy-use efficiency of automobiles.

Additionally, it should be noted that the research in this paper still has the following limitations.

(1)

The calculation of infrastructure costs does not fully reflect regional differences. In the study, the cost of charging infrastructure (such as the ratio of charging points to vehicles, and the weights of different charging methods) adopts the national average data, without distinguishing the differences in charging accessibility between urban and rural areas. Such differences can lead to deviations between the estimated cost of infrastructure and the actual situation.

(2)

The impact of regional heterogeneity of the power network on BEV emissions was not deeply explored. The research assumes that the national power structure tends to be similar, but in fact, there are significant differences in the energy structure among the provinces in China (for example, hydropower accounts for a high proportion in Sichuan, and coal-fired power generation is dominant in Inner Mongolia). Provinces that rely on coal have a higher carbon intensity in their power grids, which may weaken the emission reduction advantages of BEVs. In contrast, regions rich in renewable energy may amplify their environmental benefits. Therefore, this difference will affect the carbon emission cost throughout the entire life cycle of BEVs. Additionally, although the research proposed measures such as power decarbonization, it failed to analyze the spatial heterogeneity of the measures’ effects in combination with the differences in energy structures among provinces. For instance, in provinces with a high proportion of coal consumption, the marginal benefit of power decarbonization for BEV reduction may be more significant, while in regions with a high proportion of renewable energy, the role of other measures such as energy efficiency improvement will be more prominent. Future research can further incorporate the regional dimension to refine policy recommendations and emission reduction paths for different regions.