Electric light-duty vehicles have decarbonization potential but may not reduce other environmental problems

The study examines whether and under what conditions electric vehicles (EVs) reduce environmental impacts compared to internal combustion engine vehicles (ICEVs), hybrid electric vehicles (HEVs), and plug-in hybrid electric vehicles (PHEVs). Transportation contributes roughly 25% of energy-related greenhouse gas emissions and continues to grow. Light-duty vehicles are the largest share of transport emissions. EV adoption is rapidly increasing due to potential decarbonization benefits from zero tailpipe emissions and the possibility of charging with low-carbon electricity. However, overall life cycle impacts depend strongly on the local electricity grid mix and the specific impact category considered. Prior evidence shows EV life-cycle emissions can exceed those of HEVs or ICEVs in fossil-intensive grids. The paper aims to provide a comprehensive, multi-indicator life cycle assessment across global and country contexts (Norway, US, China) and to test how results vary with vehicle usage parameters (e.g., PHEV electric driving share) and evolving electricity mixes.

Prior LCA literature largely centers on global warming potential (GWP), typically finding that EVs reduce GHG emissions except in coal-dependent regions. Studies report BEVs achieving 32–50% GHG reductions in China, the EU, and the US, despite higher production-phase emissions. Grid mix explains up to ~60% of EV lifetime carbon emissions variability, and static grid assumptions can overstate benefits. Health-related impacts, especially respiratory effects (RE) from PM2.5, are less explored. Some studies find EVs reduce CO2, VOCs, and NOx but may increase PM2.5 and SO2. Environmental justice analyses show EV-induced emissions can be shifted from urban users to rural, lower-income areas near power plants or fuel production sites, potentially exacerbating disparities; for example, poorest regions in China may bear 40% greater pollutant exposure burdens than wealthier areas. This motivates multi-impact, geographically specific analysis beyond GWP alone.

The authors conduct a cradle-to-grave life cycle assessment (LCA) following ISO 14040/14044, covering construction, operation, maintenance, and end-of-life for four medium-sized light-duty vehicle types: ICEV, HEV, PHEV (48.3 km all-electric range), and BEV (482.8 km range). A 320,000 km lifetime driving distance is used. Modeling is performed in openLCA 2.1 using Ecoinvent v3.7 (cut-off) and GREET 2022 for life-cycle inventory, with impact assessment via TRACI 2.1. Electricity mixes are region-specific and include current scenarios and 2030 projections under IEA World Energy Outlook 2023 Stated Policies Scenario (STEPS) and Announced Pledges Scenario (APS) for the US and China. The study separates the total life cycle impact into a vehicle cycle (production, maintenance, disposal; treated as an intercept β) and a fuel cycle (well-to-tank, tailpipe combustion, and electricity generation; treated as slope α multiplied by lifetime distance). Key parameters affecting α include: regional electricity mix, local emission standards, vehicle fuel economy, gasoline grade, PHEV electric driving share (Utility Factor), and driving distance. The analysis maps which vehicle minimizes impacts over a 2D landscape of lifetime distance vs. PHEV electric driving share for each region and scenario. Uncertainty analyses adjust GREET defaults to reflect ranges for medium cars (e.g., ICEV and HEV fuel economy, BEV electricity consumption, PHEV electric driving share 0–100%) based on the 2024 Fuel Economy Guide and literature. The 2030 grid is used as a mid-point representation for vehicles entering service around 2024 with ~12-year lifetimes. Results focus on GWP and respiratory effects, with additional categories in Supplementary material.

• Electricity generation impacts per kWh (selected values from Fig. 1):
  • GWP (kg CO2-eq/kWh): Global current 0.724; Norway current 0.023; US current 0.548; US 2030 STEPS 0.241; US 2030 APS 0.181; China current 1.037; China 2030 STEPS 0.579; China 2030 APS 0.517.
  • Respiratory effects (kg PM2.5-eq/kWh): Global current 1.1e-4; Norway current 1.9e-5; US current 1.3e-4; US 2030 STEPS 2.3e-4; US 2030 APS 1.4e-4; China current 8.8e-4; China 2030 STEPS 4.9e-4; China 2030 APS 4.3e-4.
• Vehicle cycle impacts (Fig. 2): BEVs have the highest vehicle-cycle GWP and RE; PHEVs lower than BEVs, HEVs and ICEVs similar. BEV vehicle-cycle impacts are driven by batteries (≈50% of GWP; ≈45% of RE). BEVs exceed PHEVs by ≈44% (GWP) and ≈33% (RE); PHEVs exceed HEVs by ≈18% (GWP) and ≈13% (RE). Maintenance burdens are lower for BEVs due to no engine oil.
• Current-scenario GWP landscapes (Fig. 3):
  • Global (0.72 kg CO2-eq/kWh): BEVs minimize GWP only for high lifetimes (>270,000 km) and low PHEV electric shares (~26–31% threshold). Elsewhere HEVs dominate at low distance/low e-share; PHEVs elsewhere.
  • US (0.55 kg CO2-eq/kWh): BEVs outperform HEVs above ~82,000 km; BEVs beat PHEVs when PHEV e-share is ~27–70% depending on distance.
  • Norway (0.02 kg CO2-eq/kWh): BEVs dominate; surpass HEVs above ~26,000 km and PHEVs even up to ~94% PHEV e-share, depending on distance.
  • China (1.04 kg CO2-eq/kWh): HEVs strongly dominate; BEVs and PHEVs fail to offset battery-related burdens due to high grid GHG intensity.
• 2030 GWP landscapes (Fig. 4):
  • US STEPS (0.24): BEVs dominate; beat HEVs above ~36,200 km and PHEVs up to ~89% e-share. US APS (0.18): thresholds improve to ~32,600 km vs HEVs and ~90% vs PHEVs.
  • China STEPS (0.58): BEVs beat HEVs at ~90,600 km and PHEVs up to ~67% e-share. China APS (0.52): ~70,500 km vs HEVs and ~74% vs PHEVs.
• Respiratory effects (current; Fig. 5): In Global, US, and China, after ~9,200–10,000 km, HEVs have the lowest RE because electricity production RE exceeds gasoline well-to-tank and tailpipe RE; Norway is the exception where BEVs minimize RE due to hydropower.
• Respiratory effects (2030 STEPS/APS): Patterns persist; HEVs continue to dominate RE in US and China.
• Tailpipe vs non-tailpipe RE at 320,000 km (Fig. 6): Tailpipe RE is <4% of total. Non-tailpipe sources (electricity generation for EV driving; well-to-tank for gasoline) dominate. Under APS, BEV and PHEV RE decrease vs STEPS due to cleaner grids, but totals remain higher than HEVs (HEV totals ≈10.3 kg PM2.5-eq in the US and ≈10.1 kg in China). For US STEPS, increasing PHEV e-share from 0% to 100% raises total RE from ~10.8 to ~18.1 kg PM2.5-eq.
• Overall: EVs have strong decarbonization potential as grids decarbonize, but may not reduce other impacts, especially RE, and can shift and concentrate pollution burdens in rural/low-income communities.

The results show that EV environmental performance is contingent on grid composition and impact category. For GWP, BEVs increasingly outperform HEVs and PHEVs as grids decarbonize (e.g., Norway; US 2030 STEPS/APS). In fossil-intensive grids (e.g., current China), HEVs minimize GWP. For respiratory effects, HEVs generally minimize impacts in the US and China because electricity generation contributes substantial PM-related burdens; Norway is an exception. Splitting RE into tailpipe versus non-tailpipe sources reveals that most RE arises away from urban roads—from electricity generation and petroleum supply chains—implying electrification shifts, and can sometimes amplify, pollution in communities near power plants, often rural and lower-income, raising environmental justice concerns. PHEVs with high electric shares can have lower GWP and RE than BEVs due to smaller batteries and lower electricity consumption; however, benefits depend on charging availability and driving patterns. Policy implications include: tailoring incentives to regional grid cleanliness (e.g., favor HEVs in dirty grids short term; accelerate clean electricity deployment), expanding charging infrastructure to increase PHEV electric shares while addressing range anxiety without oversizing batteries, encouraging early retirement of high-emitting vehicles, extending EV battery lifetimes, and mitigating power plant emissions with attention to exposure in vulnerable communities.

EVs are not inherently clean across all metrics; their benefits strongly depend on electricity grid cleanliness and usage patterns. While EVs can substantially reduce GWP as grids decarbonize, they may not provide concurrent reductions in respiratory effects and can shift pollution to rural, lower-income areas, potentially worsening environmental injustices. PHEVs and HEVs can serve as transitional solutions in regions with higher-emission grids. Future work should refine geographic resolution (state/province-level grids), incorporate battery end-of-life pathways (re-use, recycling), analyze charging behaviors (including fast charging and alignment with renewable availability), assess grid-support solutions (storage, vehicle-to-grid), expand charging infrastructure, and integrate economic analyses with LCA.

The study excludes battery end-of-life due to data scarcity, despite its potential to significantly affect results via second-life and recycling pathways. Projections of future grid mixes (IEA STEPS/APS) introduce uncertainty. Available datasets constrain the dynamic and spatial resolution (country-level vs state/province-level), and LCA impact results differ from actual exposure patterns for pollutants. Several parameters (e.g., fuel economy, PHEV electric share, BEV consumption) are uncertain and varied in sensitivity analyses but still may not capture all real-world variability.