New research on emissions strengthens case for a 20mph default urban speed limit

New research[1] from engineering consultants, Future Transport, models the impact of capping speeds at 20mph vs. 30mph. This “real life” modelling that takes account of the stop/start nature of urban traffic yields a very different result from traditional steady-state models. It shows significant and substantial reductions in emissions: CO2 lower by 26% and NOx 28% lower. With UK hosting COP26, campaigners are calling on governments to set 20mph or 30km/h limits as national urban/village defaults.

Although the auto industry is fully aware of the impact of acceleration on vehicle emissions, it does not publish the results. Basic physics means that 2.25 times more energy is required to reach 30mph than 20mph. When this is repeated in the real-world environment, where we slow down at junctions, crossings, congestion points and other hazards, acceleration becomes the dominant factor in overall journey emissions.

Future Transport modelled the CO2 and NOx emissions for accelerating from stationary to between 5 and 50 mph for a number of vehicles, with the following results for a petrol Ford Focus.


Repeated acceleration and braking, rather than steady-state, represents a far better modelling of real-world emissions in our congested cities and towns.

Wales is already planning to change the national urban default limit of 30mph to 20mph by 2023. Besides “liveability” and casualty reduction, transport carbon reduction[2] is also cited as a reason to change. The Scottish Government[3] has announced its plans to set 20mph as a norm across the country from 2025.

Speed limit reductions are a key initiative to reduce climate-warming CO2 emissions and harmful NOx. They also have a significant effect on public health through air quality improvements and active travel.

Rod King MBE, Founder and Campaign Director for 20’s Plenty for Us commented

“It is clear that repeated acceleration dominates emissions in town driving. This research quantifies the effect and shows how reducing maximum speeds can have a significant beneficial effect on emissions. It’s time for all governments to say 20’s Plenty for the planet and for our health. With COP26 approaching it is an effective step towards transport carbon reduction”




Open PDF

Showing 16 reactions

Please check your e-mail for a link to activate your account.
  • Tim Coote
    commented 2023-05-09 08:00:59 +0100
    @Sam Bailey, I don’t think that you’ve got a firm grip on how to measure, mitigate and ameliorate risk. In the context of road safety, a good starting point is the micromort (one in a million chance of dying from a particular activity), which gives a good framework for comparing different risks. It’s a bit of a blunt instrument (being based on averages, you have to think of the relevant population). However, it’s an approach that’s been developed by statisticians, and I can recommend David Spiegelhalter’s videos on the risk/value trade off of eating a bacon sandwich, or going cycling (DS is an ex-president of the Royal Statistical Society, and Winton Prof of Understanding at Cambridge University). ( In the context of pedestrian and driver safety, here’s a ready reckoner:, which shows the risk to drivers and pedestrians, which can easily be compared to ‘just living’.

    Looking at the ‘ready reckoner’, car users are more at risk from traffic than pedestrians. I don’t think that the need for a crash cage stands up to much scrutiny. Not included in the list is the risk of cycling, which, from memory is ~1 micromort per 20 miles. However, overall, cycling reduces loss of QALYs from fitness degeneration to such an extent that time spent cycling is ‘free’: 15 mins cycling extends your life by 15 mins, even accounting for the risk of an accident.

    To take a worked example, I live in a town of 18,000 souls. So the background death rate is around 180 per year. Between 1979 and 2021 (last year of available figures), there were 25 road deaths in 30mph zones, of which 12 were pedestrians. So in the 42 years, there were ~7,500 deaths of whom 12 were pedestrians (0.2%). It is also worth noting that the death rate (and other road accident rates) continues to fall at ~2.5% pa. Is that a big incremental risk? Is there more risk to pedestrians in accepting a lift in a 30mph zone (which is not considered high risk, I don’t think), than walking?

    If there is a lot of traffic, as there is at some times of day, then it makes sense to keep the streams separate with specific crossing points whatever the speed limit (most accidents to pedestrians seem to happen where the traffic speed is well below 30mph). If there is not a lot of traffic, then it is not necessary to find a crossing – although one may wish to do that. Children can walk to school at lower risk than being at school: they are very much safer now than they used to be. It may well be that the biggest traffic risk is from the parents of other children ferrying their charges around.

    A QALY (Quality Adjusted Life Year) is not a +ve or -ve thing, it’s a measure that can be used to give finer detail when looking at the frequency/impact of hazards.
  • Sam Bailey
    commented 2023-04-21 15:43:35 +0100
    @tim coote. I should qualify the last point further – the driver gets an instantaneous positive benefit at the point of driving faster. Over the long term though, the driver suffers from the same negatives as the rest of society.
  • Sam Bailey
    commented 2023-04-21 14:22:41 +0100
    @tim coote. There is a cost to the rest of society of you driving faster. Even if you don’t hit a pedestrian, there is a cost to them of the additional risk that you have exposed them to – financial markets price this, and charge a lot for taking risk. The pedestrian has no way to charge you for it, though you pay some of it through your insurance. There is very substantial cost to society of taking the necessary mitigation factors to minimise the risk that your faster driving exposes them too – they can’t cross the road, or have to walk further to get to a pelican crossing. They can’t let their children cycle or walk, so have to drive them. Effectively at 30mph the road becomes only usable to people in a crash protection cage, so everyone has to take a car, which costs them money. As well as the additional pollution, faster traffic causes substantially more noise, which blights the road, reduces the value of the property and the quality of life of people who live on it and people who travel down it. This in turn encourages sprawl, as people want to live further away from faster roads, which then creates more traffic. So yes, people are harmed by it. With regard to the QALYs – for the person driving they are all positive, and the person not driving they are all negative. Why should the driver get all the positives?
  • Tim Coote
    commented 2023-04-21 12:47:45 +0100
    @Sam Bailey wrt comment on journey times. I think that you’re missing the point: average journey times don’t matter. If luck were with him – say at 99th centile (or he were on a non-urban road, early in the morning) then with a 20mph speed limit 4 miles takes 12 mins, but with a 30mph limit, it’s possible that he could be there in 8. Why would he want to extend his aspirational time by 50%? Who is harmed in any way apart from him?
  • Tim Coote
    commented 2023-04-21 12:33:03 +0100
    @sam bailey, thanks for pointing out the energy conversion error. That was sloppy of me. However, the other numbers come from the graphic in the article, rather than the original paper. Looking at the paper, it’s a simulation, based on inner London traffic-light density/road layout and unrealistic acceleration/deceleration characteristics. If the traffic choke points are 160m apart, a typicl driver may well not get above 20mph and wouldn’t even try (contrary to the model assumptions). That’s not a reason to have a 20mph speed limit when the traffic is flowing freely. We do need some actual measurements of the fuel consumption and driver behaviours.

    The paper seems to follow the assumption that ‘average speed’ is a meaningful measure for travel. It is for some situations, but not many. The main roads that I had trouble with were Quarry Hill Rd, Brook St and Upper Haysden Lane. When entering Quarry Hill Rd, the limit drops from 70 to 30, so there is no acceleration. It is quite common for the interpretation of 20mph limits to put the zones in the middle of 30mph zones.

    I quite like the pollution comments in the paper, although they are simplistic. I wonder if you added up the lost QALYs from actual pollution caused by vehicles and subtracted the time wasted travelling too slowly in 20mph zones whether the answer would be +ve or -ve.

    STATS19 data show that accident victims, aside from cyclists mostly occur near junctions. Often these junctions are poorly laid out, either causing the confusion between actors or slowing the traffic to a crawl before any collision, or both.
  • Sam Bailey
    commented 2023-04-20 16:00:40 +0100
    @Chris Bridgland.
    A reduction of maximum speed from 30mph to 20mph doesn’t result in a 33% increase in travel times.
    If I want to drive the 4 miles from Kingston to Richmond along Richmond Road at 30mph, it should take 8 minutes. But if I type the journey in to Google Maps it tells me even with light traffic it will take 16 minutes. That’s because 8 mins will be spent waiting at traffic lights. You can therefore drive the distance between the lights at a lower speed, maintain the position in the traffic queue at the lights, and still get to your destination at the same time. Occasionally you might miss a green light, but the probability is quite low. The traffic simulations in the paper model this effect. The average speed increased by only 9% going from 20 mph top speed to 30mph top speed, despite it being a 50% increase in maximum speed.
  • Sam Bailey
    commented 2023-04-20 13:43:21 +0100
    Tim – there are some important errors in your maths. You appear to have assumed a vehicle mass of 1000kg. Very few cars on the road weigh 1000kg. A Nissan Micra plus driver is mare than 1000kg. A Ford Focus plus driver is around 1400kg. SUVs are around 1700-2300kg.
    The energy to accelerate the rotating masses in the flywheel and drivetrain is also significant – they have to be repeatedly sped up with each gear change.
    Your maths assumes that the energy in the petrol is converted to mechanical energy with 100% efficiency. An IC engine will be around 20-30% efficient. The paper uses the efficiency maps from real engine tests, and shows the (chemical) energy for a Focus to reach 30mph to be 230KJ more than to reach 20mph, not the 50KJ (mechanical) energy you calculate.
    You have to allow for the distance driven whilst still accelerating and braking – the whole distance is not covered at the ‘cruise’ speed.
    The paper calculates the distances that must be travelled without decelerating and accelerating for the different vehicle types for 30mph to be more efficient than 20mph including all these effects. For the Petrol Focus it is 0.91km, for the Mazda CX-9 it is 1.3km, and for the diesel X5 SUV it is 3.08km.
    I don’t believe there are many urban road sections in the UK where you can travel 1km every time without slowing down for a junction, a vehicle turning in front of you, a pedestrian crossing or similar. If I do a Google Maps view of Tonbridge, I can see one section of the A26 Cannon Lane where I might get to drive 1km without hitting a junction. So if there’s no other traffic turning on and off the road, and I’m in a smaller petrol vehicle, the CO2 will be about the same at 20mph and 30mph. For every other road section I can see, and the majority of vehicles on the A46, the emissions will be higher at 30mph than 20mph.
    Note that Cannon Lane is a 40mph limit – in which case the emissions will be higher for a vehicle doing 40mph than one doing the same road section at 20mph, as the distance required to travel between stops for 40mph to be more efficient than 20mph for the Focus is 3.18km, and is even longer for the larger vehicles.
  • Tim Coote
    commented 2023-04-19 17:45:49 +0100
    fwiw, it is worth noting that fuel efficiency of EVs falls with speed (it’s an inverse quadratic relationship as air resistance exceeds a relatively stable rolling resistance).

    Looking at the ICE effect, 50KJ energy difference to get to 30mph would take ~100m for the faster car to be more efficient overall. That may be realistic in parts of London and the centres of some big cities, but not for most towns most of the time where it’s normal to travel for up to 1Km without having to brake. (energy density of petrol: 46MJ/L, energy efficiency differential between 30km/h and 50km/h ~1L/100km.)
  • Rod King
    commented 2022-10-03 19:51:52 +0100
    Thanks for your comment Tim. Its always useful to have a different perspective.
  • Tim Coote
    commented 2022-09-22 11:30:30 +0100
    The paper isn’t really research, it’s an exercise in modelling/simulation – something that I used to do professionally. Although the paper describes some model validation, this is limited to vehicle performance, rather than actual driving patterns. As a modelling exercise, it is intended to try to simulate ‘typical’ London traffic, although it is not clear whether that’s a typical period of time, or driving experience. I would expect any modelling that elaborates a ‘typical’ scenario to also provide a credible interval (to use a Bayesian term), as do the IPCC climate change scenarios. It’s very debateable as to how typical typical London traffic conditions are for most urban driving. As there is no relevant validation, it is not possible to know how representative the results would be for actual driving.

    The results are not validated at all against actual driver behaviour. Some of the discussion is wrong, and, rather typically, shrouded in unnecessarily complex maths that is simultaneously over-simplistic (e.g. the discussion on how far a vehicle would have to accelerate to a higher speed to be more efficient than a car limited to 20mph.

    There is value in educating drivers on how to drive efficiently. However, it’s debatable whether the best overall/ethical outcome arises from imposing regulations on peripheral characteristics (the speed limit, all the time), rather than, say appealing to the drivers’ pocket.

    For the most part, the environmental and health impacts of driving result from specific locations where road layouts are not suitable, rather than large areas (e.g. look at the locations of STATS19 incidents to make this very evident from an accident point of view). And technology continues to improve both the impact in both dimensions.
  • Rod King
    commented 2022-08-22 10:52:35 +0100
    Thank you Chris

    I will bypass your first two “firstlies”, but if I can comments on your “secondly” (or is it “thirdly”), the emissions in the article are measured in grms/km. Hence a reduction of 25% per km is also a reduction of 25% per trip.
  • Chris Bridgland
    commented 2022-08-15 13:43:55 +0100
    Firstly, I do understand the need for speed limits where is makes sense and Cities and commercial towns need to consider this carefully, but there are 2 items that aren’t addressed.

    Firstly, much of this is about education. I grew up with the Green Cross Code Man, Charley Says series and films about things such as playing in a farm. The lack of such teachings like these on our school curriculums, televisions and radios, to be replaced by beauty products, bladder control and erectile dysfunction advertising is such a missed opportunity for our government.

    Secondly, this research misses the effect on increased travel times and associated stress levels that can impact some people in a services environment, such as carers who need their vehicles to get around, or Commercial/HGV delivery drivers who need to complete their schedules. A 25% reduction in emissions but a 33.3% increase in travel times. The public transport in this country is still too inconvenient for many to consider as an alternative, as some councils, such as Kingston-upon-Thames, use this argument as one of their reasons for the breadth of their 20MPH limits that they have applied.
  • Rod King
    commented 2022-07-14 20:52:42 +0100
    Thank you Andrew. I assure you that the researchers are very well acquainted with the design of modern cars, but would challenge the idea that they are designed to run in top gear at 30mph. I am not sure where you obtained that information. And of course the whole point of gears is that you can run at a slower speed and maintain the same revs. That is what gearboxes do. And engine management systems meter fuel to the engine based on load and revs in order to burn as leanly as possible.
  • Andrew Maskell
    commented 2022-07-12 13:30:05 +0100
    Does your research take account of the fact that modern cars are deigned to run in top gear at 30 miles perhour but have to change to a much lower gear at 20 mph, thereby using more fuel and therefore producing more pollution? I think your conclusion is too simplistic and wrong.
  • Rod King
    published this page in Briefings 2021-10-05 22:08:13 +0100
  • Rod King
    published this page in Briefings 2021-10-04 19:58:59 +0100