Electric cars vs petrol engines: which pollutes more over a lifetime? Manufacturing, grid mix, and the latest lifecycle numbers

Comparing ‘pollution’ from electric cars and conventional engines requires agreeing on what counts. Tailpipe-only accounting flatters petrol and diesel: it ignores oil extraction, refining, shipping, and methane leaks upstream of the pump, just as it would ignore power-plant stacks if one only looked at a charger cable. Serious comparisons use lifecycle assessment (LCA): manufacturing (including the battery), maintenance, fuel or electricity supply, and driving over a realistic lifetime and mileage. The question becomes not ‘which car smokes at the stoplight?’ but ‘which chain of industry warms the climate and harms health more per kilometre over twenty years?’

On greenhouse gases—the dominant long-lived pollutant tied to transport climate impact—recent modelling from the International Council on Clean Transportation (ICCT) gives clear EU-scale numbers for medium-segment cars sold in 2025. Using a projected 2025–2044 average European electricity mix, ICCT estimates battery electric vehicles (BEVs) at about 63 grams of CO₂ equivalent per kilometre (g CO₂e/km) across the full lifecycle. The comparable sales-weighted average for gasoline internal combustion vehicles, including biofuel blending in the fuel pool, is about 235 g CO₂e/km. That is roughly a 73% lifecycle emissions reduction for the BEV. Diesel sits essentially tied with gasoline in the same analysis (about 234 g/km), which surprises readers who associate diesel only with efficient engines but forget higher fuel-production burdens and real-world usage patterns used in the model.

Manufacturing is where sceptics focus, and the data supports part of their point: ICCT estimates BEV production emissions—including battery cell and pack supply chains—at roughly 40% above those of a comparable ICE vehicle. That penalty is real: cathode chemistry, refining, and energy-hungry cell plants concentrate emissions early in the vehicle’s life. The same report, however, calculates that those extra manufacturing emissions are paid back after about 17,000 km of driving—typically one to two years for a household car—because the use phase dominates for ICE cars once you include fuel supply chains. After payback, every additional electric kilometre on a decarbonising grid widens the gap.

The EU story improved rapidly: ICCT notes today’s BEV lifecycle estimate is about 24% lower than its comparable 2021 study, chiefly because the electricity mix is projected to keep shedding coal and because manufacturing efficiency assumptions moved. If the same BEV were charged on 100% renewable electricity, the study places lifecycle intensity near 52 g/km—about 78% below the gasoline baseline—showing that grid cleanliness is the master dial after vehicle efficiency.

Across the Atlantic, ICCT’s July 2024 U.S. brief on model-year 2024 sedans and SUVs finds a parallel picture with American grid carbon intensity baked in: BEV sedans post about 66–70% lower lifecycle greenhouse emissions than conventional gasoline vehicles; BEV SUVs about 71–74% lower, with the exact percentage depending on how carbon-heavy the regional grid is modelled. The U.S. Environmental Protection Agency has cited Argonne National Laboratory work finding roughly a 60% lifecycle reduction for a 300-mile-range BEV versus gasoline when manufacturing, fuel cycles, use, and disposal are included—same family of conclusion, slightly different vintage and assumptions.

Deduction for climate pollution over a typical lifetime: for average new cars in Europe or the United States under mainstream 2024–2025 LCA assumptions, the battery car still wins on CO₂e even though it starts ‘dirtier’ at birth. The ICE vehicle loses on cumulative fuel-cycle emissions; the BEV loses on upfront industrial emissions but catches up quickly and pulls away. The exceptions are instructive: if analysts shrink lifetime mileage, freeze the grid at today’s coal-heavy regional mixes without improvement, or ignore real-world fuel consumption gaps that favour underestimating gasoline use, spreadsheets can temporarily flatter combustion. ICCT explicitly warns that using a static instead of improving electricity mix, shortening assumed vehicle life from about twenty years, or swapping type-approval fuel figures for real-world driving distorts comparisons—usually in favour of ICEVs.

Greenhouse gas is not the only pollution people breathe. For urban nitrogen oxides, volatile organics, and fine particles directly emitted where children walk to school, BEVs have zero tailpipe discharge; benefits concentrate in cities even when power plants remain fossil-fired—because those plants are often farther from dense populations and increasingly filtered. Non-exhaust particulate matter complicates the moral: heavier BEVs can increase tyre and road-wear dust, while regenerative braking cuts traditional brake dust. Net health outcome depends on weight, tyre compounds, driving style, and baseline air quality; lifecycle GHG studies rarely capture those aerosols fully, so honest reporting treats climate and toxic air as overlapping but not identical scorecards.

Readers hunting investment or purchase implications should treat any single number as scenario-bound: battery size, SUV mass, local kWh carbon intensity, and how long the first owner keeps the car all move the breakeven odometer. What the peer-reviewed and grey-literature consensus has converged on is directional: in representative OECD grids trending cleaner, lifecycle CO₂e favours electrics for comparable segments; manufacturing is higher for BEVs; payback distances are modest compared with total lifetime kilometres; and the gap widens as grids add renewables and factories decarbonise.

Newsorga will update this explainer when major agencies publish revised LCA methods—particularly if the European Commission or UNECE harmonise reporting rules—or when new meta-analyses shift consensus ranges for battery carbon intensity.