Plug-in hybrids emit almost as much CO₂ as petrol cars in real-world driving, report finds

Plug-in hybrids have been sold to Australian buyers as the pragmatic middle ground: electric around town, petrol for the long haul, and — for caravanners — a tow vehicle that sips fuel on the way to the van park. New real-world data from Europe suggests the "electric around town" half of that promise is barely being kept.

According to an analysis by the campaign group Transport & Environment (T&E), plug-in hybrid vehicles registered in 2024 emitted an average of 145 grams of CO₂ per kilometre in real-world driving — six times their official test figure of 24 g/km, and only 14 per cent less than conventional petrol and diesel cars, which average 169 g/km.[1][2]

That is not a rounding error. It is a structural gap between what the laboratory says and what the road delivers — and it has implications for every buyer comparing running costs on a showroom sticker.

The data behind the numbers

The figures come from the European Environment Agency's mandatory on-board fuel consumption monitoring (OBFCM), which has required real-world emissions reporting from liquid-fuelled vehicles sold in the EU since 2021. This latest analysis draws on data from roughly 220,000 plug-in hybrids.[1]

The trend is the uncomfortable part. In 2023, real-world PHEV emissions were already five times the official figure (138 g/km against 28 g/km). For 2024 registrations the gap widened to six times — while the official average improved by 15 per cent, from 28 to 24 g/km, and carmakers claimed PHEV performance was getting better.[1][2]

Real-world emissions, meanwhile, went the other way: up 5 per cent year on year.[1]

The finding is not an outlier. The International Council on Clean Transportation reached a similar conclusion in a separate evaluation of around eight million vehicles published earlier this year, estimating real-world 2023 PHEV emissions at roughly five times official test values.[3]

Why the gap exists

The laboratory test assumes plug-in hybrids spend a large share of their kilometres running purely on battery. It doesn't. The T&E analysis found that even when the vehicle is nominally in electric mode, the combustion engine is operating for roughly one third of the kilometres driven.[1][2]

The reasons are practical rather than conspiratorial: many owners don't plug in every day, cold starts and hard acceleration wake the engine, the extra weight of carrying two powertrains penalises efficiency, and a smaller battery depletes quickly on open-road driving.[3]

The mechanics sit in the test itself. Official CO₂ ratings assume a "utility factor" — a fixed estimate of the share of driving done on battery power. That estimate was built from early adopter behaviour and charging assumptions that real-world data has progressively contradicted. Europe began correcting the factor in 2025/26 precisely because on-board monitors made the divergence impossible to ignore; the Australian figures buyers see on labels and in reviews are still the uncorrected lab values.[1][3]

The financial consequence lands on the owner, not the test cycle. T&E calculates that drivers of a 2024-registered PHEV pay around €680 a year more in fuel than the official consumption figures imply — a figure that scales directly for Australian buyers paying comparable pump prices for a vehicle whose sticker economy was optimistic.[1]

The policy fight this has triggered

Europe has already moved to correct the accounting. The EU's "utility factor" — the assumed share of electric driving in official emissions calculations — was revised for 2025/26, and the effects are visible in the product cycle. T&E's analysis of the 20 top-selling PHEV models (representing 51 per cent of the 2025 market) found manufacturers responded to the correction by fitting larger batteries and longer electric ranges: the Volkswagen Tiguan's electric range more than doubled from 55 km to 119 km, keeping its official rating at 36 g/km. Across five VW models, engineering upgrades offset around 90 per cent of the emissions increase the correction would have caused.[1]

The next correction is due to begin in November 2026 — and it is politically contested. Germany is reportedly pushing to freeze the utility-factor updates as part of a broader France-Germany automotive package, with the European carmaker lobby arguing the revisions go too far. T&E's position is that even the corrected factors still understate real-world PHEV emissions by around 56 per cent, and that freezing them would reward cars that pollute nearly as much as conventional combustion vehicles while claiming an 83 per cent improvement on paper.[1][3]

There is a second-order effect that reaches Australia regardless of what Brussels decides. Low official PHEV emissions feed directly into manufacturer fleet-CO₂ targets — the looser the accounting, the fewer pure electric cars a maker is obliged to sell to comply. Stringent real-world accounting tightens that constraint in Europe, and because Australia is a small right-hand-drive market that largely inherits the product decisions of larger regions, the European compliance math influences what gets engineered, homologated and shipped here. The utility-factor fight is, in part, a fight over how much genuine electrification the global product pipeline must contain.

What this means for Australian buyers

Australia has no equivalent correction process, and the same WLTP-derived laboratory figures that European regulators are now adjusting sit behind the official numbers on Australian fuel-consumption labels. The test-methodology gap is not a European problem; it is a lab-versus-road problem, and it travels with the car.[2]

That matters most in the segment where PHEVs have made their strongest Australian case: towing. The plug-in hybrid ute and SUV category — vehicles like the BYD Shark 6, the GWM Cannon Alpha and the Mitsubishi Outlander PHEV — is marketed squarely at buyers who want to tow a caravan or camper without the range anxiety of a battery-electric tow vehicle.[2]

The real-world data doesn't kill that use case, but it does sharpen the buying question from "what's the combined-cycle figure?" to "will I actually charge it?" The honest summary from the data:

• If you can charge at home or at your destination every day, a PHEV can deliver a genuine share of electric kilometres — and the EU corrections show manufacturers will build longer electric ranges when the rules demand it. If you can't charge regularly, the 145 g/km figure is the honest baseline: you are driving a petrol car with a heavy battery, getting most of the weight penalty and little of the electric benefit. The official label's savings arithmetic will not apply to you. If you're towing, expect the combustion engine to work hard regardless of charge habits — the added mass and drag push the engine into operation far more often, which is precisely the regime where real-world figures diverge furthest from the lab.

None of this makes plug-in hybrids a poor product. It makes them a conditional product — one whose value depends almost entirely on the owner's charging discipline, something the laboratory test has never been able to see.

It is also worth separating the finding from its neighbours. The same monitoring regime that exposed the PHEV gap gives buyers a genuinely new tool: official emissions claims across all drivetrains are becoming auditable against on-road data, and the direction of travel for regulators is toward real-world accounting everywhere. Battery-electric vehicles carry no such tailpipe gap by construction — their energy use is measured at the meter. The PHEV story, properly read, is less "electrification doesn't work" and more "electrification that depends on owner behaviour has to be designed and sold honestly." Buyers who understand that distinction are better placed than buyers handed a lab figure.

A buyer's checklist before the PHEV pitch works on you

The sales conversation around plug-in hybrids leans heavily on the official figures. Given what the on-road data now shows, it is worth walking in with a short list of your own:

1. Where will it charge? A home wallbox or reliable workplace charger is the difference between a PHEV that runs electric most days and one that is a petrol car with a dead-weight battery. If your answer is "the occasional public charger," the real-world emissions line applies to you. How long is the electric range, honestly? Check the real-world range reported by owners and independent testers, not the WLTP figure. Short electric range is the single biggest predictor that the engine will carry most of the load — it is also what the EU's correction has already forced manufacturers to extend. What does it do uncharged? Ask for the fuel consumption figure with an empty battery. That number — not the combined cycle — is the honest cost of the kilometres you'll drive when towing, road-tripping or simply forgetting to plug in. What is the resale case? As more jurisdictions move toward real-world accounting, vehicles bought on optimistic lab figures face the same reassessment the EU is now performing. A PHEV purchased for its label economics may not hold that story at trade-in time.

The bottom line

The 2024 data closes the loop on a debate that has run since PHEVs first appeared: whether they are a bridge to electric driving or a way to keep combustion engines on the road with green credentials. The answer, at current charging behaviour, is closer to the second — 14 per cent better than a petrol car is not the 83 per cent improvement the label implies.

For Australian buyers, the practical takeaway is simple. Read the real-world number, not just the sticker. Ask how the vehicle performs when it isn't plugged in — because for a meaningful share of kilometres, that is exactly what you're buying.

Sources

• https://www.transportenvironment.org/articles/plug-in-hybrids-emit-six-times-on-average-what-official-tests-claim-new-eu-data https://thedriven.io/2026/10/19/plugi-in-hybrids-emit-almost-as-much-co2-as-petrol-and-diesel-cars-report-finds/ https://www.electrive.com/2026/10/16/plug-in-hybrids-emit-six-times-more-co2-than-officially-stated/