Jeep Grand Cherokee

Why wait til 2035 to ban the most polluting new cars?

Background: “[The UK] Government is seeking views on bringing forward the end to the sale of new petrol, diesel and hybrid cars and vans from 2040 to 2035, or earlier if a faster transition appears feasible.” This post is based on my response to that consultation.


The electrification of road transport is a crucial part of reducing greenhouse gas emissions. It would also slash air pollution, which has been linked to a wide range of harms. Most recently, there has been strong suspicion that air pollution both worsens COVID-19 and helps viruses to travel further (though more work is needed). But even before the current crisis, 40,000 deaths per year in the UK were attributable to air pollution. It’s linked to asthma, miscarriages, dementia, and even day-to-day cognitive harms. Indeed, it’s possible that “…phasing out gasoline and diesel-burning vehicles and reducing the burning of coal and natural gas in power plants could have wide-ranging benefits for labor productivity, school performance, and the long-term care needs of the elderly.” What’s more, many of the same vehicles that are the most polluting are also the most physically dangerous. And on top of all this, reducing the use of biofuel is the easiest way to reduce demand for palm oil, preventing further deforestation; while reducing fuel imports makes UK living standards and security less dependent on oil-rich regimes.

The solution – switching to electric vehicles (EVs) – is perfectly achievable. As well as being nicer to drive (with no gears necessary), EVs are already cheaper than non-EVs, on a lifetime basis. Even the up-front cost is expected to be lower “by the mid-2020s”: maybe even before the next election. But every year matters in terms of greenhouse gas emissions. And the faster the UK scales up, the more rapidly costs will fall and technologies further develop, which will then help the entire world’s vehicle fleet transition.

So we can do better than ending the sale of new petrol and diesel cars and vans by 2035. We can’t tell future generations we really cared about the “climate emergency” – or any of the other concerns discussed above – but decided not to restrict even new polluting cars in this 5-year parliament (2019-24), nor the one after (2024-29), nor the one after that (2029-34). By contrast, in Norway pure petrol/diesel sales are already down to 20% of new sales, and sales will end by 2025. We should aim for a date of 1 January 2030 at the very latest. But below I want to set out a slightly different approach.

The most polluting new cars should be phased out far sooner

The debate surrounding this policy has discussed the ban in terms of a single number: whether the ban comes in 2040, 2035, 2032 or 2030. But it doesn’t need to be a single date: it could differ for different vehicles.

The key argument against banning all new petrol and diesel cars very soon (say, in 2022) is that the world can’t produce enough EVs that quickly, despite a phenomenal pace of growth. When not in lockdown, the UK consumes 2 million new cars a year, which seems to exceed current global battery EV production. Even if the government announced it wanted a rapid transition, new cars take years to design, plan, build and distribute: as do massive battery factories; and as do lithium mines (not to mention the process of developing new battery technologies). At the same time, the country would need to greatly ramp up the infrastructure for electricity generation and distribution. So if we ended the sale of all ICE cars in 2022 then there would be a massive shortfall in new cars, presumably pricing more people out of car ownership.

That is a reasonable (albeit fairly conservative) argument against banning all new ICE cars soon. But none of that is an argument against substituting out dirtier new ICE cars for cleaner new ICE cars. That substitution would not lead to any shortage of cars, nor great new pressures on infrastructure. So we should ask: given the climate emergency, and harms of air pollution, why should particularly polluting new gas-guzzlers still be on sale in the UK in 2029? I don’t think there’s any good answer to that question.

It’s clear that not all ICE cars are the same in terms of environmental harms. The median model of new car in 2020 produces 166g of CO2 per km (using WLTP figures), while the worst offenders – the Rolls Royce Ghost, Rolls Royce Dawn and Chrysler Jeep Grand Cherokee (pictured) – emit over 380gCO2/km. This means that you can actually reduce emissions more by replacing the most polluting new cars with medium-polluting ones than by replacing a medium-polluting car with an EV. That’s not an argument against moving to 100% EVs, but it does show that policy can make a big difference even before implementing a complete ban.

To show the full picture, the chart below ranks every model of new car registered for sale in April 2020, from those with the lowest emissions (EVs) on the left, through to the extremes of pollution on the right (using Vehicle Certification Agency data).

So why not have a rolling ban, that starts off by ending the sale of the most polluting new cars? Below is one example of such a phased-in approach: beginning in 2022 by ending the sale of cars emitting more than 300gCO2/km. (For vans the approach could be similar but slower, but I’m focusing here on cars.)

To give a rough indication of what effect that would have in each year, the chart below uses the vehicle emissions data shown above to assess what proportion of car models on sale in 2020 would be affected by such thresholds. With a 260gCO2/km cap in 2023, only the most polluting 5% of 2020 models would be affected. But by 2027 the range of cars on offer would have to be very different, with 54% of 2020 models no longer allowed. By 2029 the cap would mostly only allow hybrids or EVs (at 120gCO2/km), while in 2031 sales would essentially be restricted to plug-in and battery EVs (80gCO2/km), and then finally battery EVs only from 2032 (0gCO2/km).

This proposal is not impossibly radical. I don’t think it unreasonable to ask people who have the resources to buy brand new SUVs to instead buy cleaner cars (preferably EVs) in the years to come. I imagine the best counter-argument would be that particular manufacturers – including Rolls Royce, Bentley and Jaguar Land Rover – would suffer from rapid changes in emissions restrictions (and note that JLR may be about to receive a government bailout). But these companies are perfectly capable of producing cleaner vehicles (and indeed do), and already know that the writing is on the wall for ICEs (though for a while they would still be able to export high-emission models to the US, China etc.). It might be that 2022 is too short notice, but I can’t believe that any company needs – or deserves – until 2030 to change its production plans. If those companies are to thrive beyond the 2020s, they need to be planning for a 100% EV future right now.

To reiterate: while the date for a complete ban is important, and should come relatively early, the government should consider a phased-in approach, with a more ambitious date for the most polluting vehicles and a slightly more relaxed approach for the most efficient. It’s a simple idea, but one I’ve not seen discussed so far. It deserves to be, and could help reconcile the views of those who want a cautious approach (e.g. 2035) and those who want a more ambitious one (e.g. 2030 or earlier).

Motorbikes should be included

Motorbikes are not currently included in the proposed petrol and diesel ban. But they should be, and could even have the same emissions caps applied.

The data on motorcycle emissions has not been as good as for cars (though this should change with WLTP). But emissions caps of 300gCO2/km or 200gCO2/km (as in the illustration above) would affect few (if any?) motorbikes, and limits of 180gCO2/km or 160gCO2/km would still allow the vast majority, with the average emissions being only 116gCO2/km – even before considering the potential for hybrid bikes.

Maybe there is a case for a later total ban for ICE bikes than for cars. But electric motorbikes are very much a thing already. Including motorbikes in the ban would also ensure the UK does not end up in a strange position where the cleanest hybrid cars are banned while the heaviest, dirtiest ICE motorbikes are not. If the idea is that there will be a ban at a later point, why not provide certainty for industry by clarifying that now?

The treatment of ICE motorbikes should also bear in mind the issue of noise pollution. In my experience, motorbikes are one of the key sources of noise pollution: so the quicker the switch to electric motorbikes begins in earnest, the better.

Adverts for ICE vehicles should be banned

The consultation asks for views on “what measures are required by government and others to achieve the earlier phase out date”. One extra idea I think should be debated is the possibility of advertising restrictions for new ICE vehicles. The UK banned cigarette adverts from TV in 1965, and from the press in 2003. Given that we’re going to ban new ICE cars entirely, why not ban adverts (or at least TV adverts) for those cars from, say, 2025?

If the ICE car ban itself were very ambitious (as in the proposal above), restrictions on advertising might not be considered worth the (miniscule) effort. But every little helps. Every ICE car that is sold in 2029 is another ICE car that will – on average – still be on the road and polluting in 2043; when we might look back and wonder why we acted so slowly in the 2020s.

Leave a Reply

Your email address will not be published. Required fields are marked *