The number of serious collisions involving cyclists and pedestrians continues to fall in Wales, three years on from the introduction of a ‘default’ 20mph speed limit on most roads.
In September 2023, amid widespread political debate and acrimony, the Welsh government introduced a default 20mph speed limit on urban and residential roads (where most, though not all, dropped from 30mph), applying to 35 per cent of Wales’ roads and increasing the approximately 870km of 20mph roads to 13,000km.
The then-Labour administration said the measures would reduce the number of deaths and serious injury caused by motorists, as well as tackling pollution and encouraging people to walk or cycle instead of drive.
The scheme has been met with fierce criticism both before and since its implementation, with the Conservative’s shadow transport secretary claiming last year that the 20mph limits continue to “frustrate and confuse” drivers.
In response to thus criticism, in April 2024 it was announced that local councils would be able to restore the previous 30mph limit on certain roads on a case-by-case basis according to local demand.
However, new figures released today by the Welsh government, detailing the number of road traffic collisions in the second quarter of this year, appear to suggest that the 20mph limits are continuing to have a profound impact on the road safety picture in Wales.
According to the figures, from April to June 2026 there were 19 per cent fewer collisions and 20 per cent fewer casualties on all roads in Wales compared to the same three-month period last year.
Compared to the same quarter in 2023, just before the 20mph speed limit was introduced, collisions involving all road users on 20mph and 30mph are down 37 per cent, with casualties dropping by 35 per cent, a record low for a year outside the Covid pandemic, the government says.
There were 45 crashes involving cyclists between April and June 2026, with 15 of those leaving the rider seriously injured. No deaths were recorded. This marks a significant improvement compared to the same quarter in 2023, when there were 78 collisions and 28 serious injuries sustained by cyclists on Wales’ roads.
“These latest figures are extremely encouraging and provide more evidence that the 20mph default speed limit is helping to make our roads safer,” Plaid Cymru’s desputy minister for transport Mark Hooper said in a statement today.
“Every one of those figures represents someone who has avoided injury, or death, on our roads.
“I want to thank the police forces across Wales for their continued work in recording this data and keeping our communities safe.
“We will continue to monitor these trends closely, as part of our ongoing evaluation of the policy, but today’s figures are a strong sign that things continue to move in the right direction.”
Speaking to road.cc earlier this month, Hooper acknowledged that, despite the clear benefits of the 20mph roll-out when it comes to reducing casualties, not enough has been done in Wales to sell the scheme to all road users, allowing it to be “weaponised” by opponents.
“I’m a big supporter of 20mph and it’s one of those things where it’s made a real difference to safety on our roads for pedestrians and cyclists alike,” he told us.
“But the way it was put across… I don’t think we spent enough time talking about the benefits with the people of Wales, we didn’t socialise the policy. We delivered the policy then tried to socialise it after.
“It’s a really good indication for how we should go about things in the future. I’m really keen to do some work on pavement parking for example, and to do that requires us to think about things in a different way than we would have done with the knowledge we’ve got from the 20mph experience.
“I think there are big lessons learned from it, and how we talk about it. There’s been a weaponisation of active travel in some quarters who are in our Senedd and I don’t expect that to change. But I think there’s common ground to be found generally about being more active and it’s something that should be pursued.
“We’re the first Welsh government to have a minister for preventative health, which talks directly to the active travel agenda. So I think it’s just about recharacterising why active travel is important for multiple reasons.”
31 thoughts on “Serious cyclist injuries down 50% and overall road casualties continue to fall since 20mph limit introduced in Wales”
Amen! Evidence – that’s how things should work.
Pity about the concern around messaging, given that in every case I’m aware of the same positives have emerged.
But that’s politics: before “why? It doesn’t do anything for me, and sounds like I’ll lose something!” After: “we don’t want to go back”.
But even evidence of “fewer people are killed / injured, people report places become nicer” can get swamped by “they’re stealing your rights! But my parking space! But the (pre-existing) motor traffic congestion was worse on the first Monday! etc”.
Not to mention the money being spent just on pro-motoring advertising may be more than goes into active travel all told. Probably a fair fraction of that goes on just poisoning the active travel well for political gains…
@chrisonabike I say: well done to the Welsh government for first acting on the available evidence, then checking that the outcome was as expected, and not buckling under the probably unavoidable moaning. Long may they continue to implement first and socialise later. That way they set an example that governments across the UK should have the courage to follow.
@slc I wouldn’t say that the Welsh government got everything right. They applied the speed limit changes uniformly, without considering exceptions, and have since back tracked in many places. This has been a massive waste of money.
@DrG82 I think it was well implemented overall, though that is because I agree with the basic point: the speed limit where vehicles interact with pedestrians etc should be 20mph* by default: 30mph can be considered as an exception (e.g a major road with significant mitigating infrastructure, controlled crossings and so on). Some Welsh local authorities have been keener to create exceptions than others for sure.
Here in Bristol, we have had widespread 20mph limits for around 15 years. The only downside I saw is that I got used to driving at 20 and would find myself doing so where the limit was 30 in other locations, to the annoyance of drivers behind (though perhaps the relief of pedestrians in front)
*There was once a substantial UK report that found the default speed limit should be around 10mph to ensure that pedestrians are not killed by drivers. That seems a long way off.
Except that, er, they did consider exceptions, in that they specifically provided in the legislation for local authorities to apply exceptions. You could argue over how long they had to implement them, but to say they weren’t considered is just not true.
And the fact that a few roads have now changed back isn’t ‘backtracking’ – it’s authorities using the legislation as designed.
@mdavidford was about to ask them if they thought the law should also been changed to ban the LAs from doing that, so as to save money…
@DrG82 -“This has been a massive waste of money”; on the other hand not doing it would merely have been a waste of life.
@DrG82 just wait ’till you hear about the “blanket” 30mph limits it was replacing, or those “NSL” signs that just got stuck up everywhere else by default…
@DrG82 I think the LAs bungled the exceptions. They had plenty of time in the run up to the implementation date to apply exceptions but mostly frittered this time away.
Regardless, I think that the reduction in casualties brings financial savings not just to the NHS but the economy as a whole and makes it a great investment.
Only a cynic would suggest that some of them might have thought there was political capital to be had in deliberately running out the clock so that they could claim they were being ‘forced’ to put in 20 limits that ‘didn’t make sense’ and then ‘fighting back’ by taking them out again…
@DrG82 is a liar.
First off, who here said they got everything right?
“I wouldn’t say that the Welsh government got everything right.”
The article contains commentary saying they didn’t get everything right. I suppose saying something obviously true sets you up to lie later…
“They applied the speed limit changes uniformly, without considering exceptions”
You are a liar. The Welsh Government guidance contained rules that would allow councils to exempt roads from 20mph. As an example, much of my route into town stayed at 30 and 40mph because my council bothered to apply the exemptions.
I particularly enjoyed your unsupported claim of “This has been a massive waste of money”.
The article states “collisions involving all road users on 20mph and 30mph are down 37 per cent, with casualties dropping by 35 per cent”.
I’d say a one-off cost of changing road signs for a permanent 1/3rd reduction in collisions is gonna save the taxpayer a chunk of healthcare costs and owners and businesses a chunk of money not fixing car bodywork.
Nothing but contempt for you. Take your lies elsewhere.
@DrG82 The implementation costs were £32M and there has been an additional £3M spent on adjustments. It’s estimated that the savings to the NHS and other bodies in terms of fewer casualties with which to deal come to around £44M. Odd economics where a £9M net saving is a waste of money.
@Rendel Harris yeah but it could have been spent on widening a few metres of road…
Okay, but is it worth it? On the one hand, hundreds of lives will be spared, and many more people spared from life-changing injuries, even if some of those are just cyclists. But on the flip side, we have to weigh that against millions of motorists being slightly delayed. Some may have to leave five minutes earlier for work every day.
It’s a tough call.
@andystow some of them may not be delayed at all but have to miss out on the fun of accelerating then braking at the next junction. That might tip the balance.
@slc so it saves fuel and brake pads? Now we have to also weigh the lost profits of oil companies and car parts suppliers. Jobs or bonuses could be at stake.
@andystow That’s pretty much how new roads are justified. If ten million motorists save ten seconds a day over the next twenty years, and motorists time is worth a lot, then you can justify spending £372m on the new roundabout in Gloucestershire. You know, the place where the locals think that a cycle lane will destroy their life.
“Serious cyclist injuries down 50% and overall road casualties continue to fall…”
This is clearly fake news, invented by the bearded, sandal-wearing, bike-riding hippies who don’t even live in Wales.
Everyone who lives there knows that the frustration and cost to drivers of having to drive at the ludicrously slow speed of 20mph is worth so much more than a few dead cyclists.
I live in the Forest of Dean, near Wales, and we’ve been trying for quite some time to get the national speed limit roads in the statutory forest reduced to 40mph, and 20mph in towns and villages. In fact, I mentioned it in the pub tonight, and the reaction from locals was, to say the least, not enthusiastic. It is a human right to be able to drive as fast as you want, and anyone saying different is an idiot.
With Local Government Reorganisation, I was hoping that the Forest of Dean District would merge with Monmouthshire and then we could adopt the Welsh speed limits, but it ain’t gonna happen, so sadly, we have to remain in England.
These stats aren’t evidence, they show correlation. Without isolating the variables they don’t actually say that much about the impact of the change.
Speed limits are an ineffective tool when the problem is drivers not driving in an appropriate manner for the conditions and environment. Those drivers weren’t following the speed limit and/or paying sufficient attention anyway!
If that is the problem and you don’t deal with it, but you want to keep reducing injuries/fatalities, you have to keep dropping speed limits until you prohibit road use entirely.
The solution is driving standards improvement (education) and enforcement.
In wider terms but more importantly it’s better city and road design, and reducing commuting distances by reforming the housing market (to an easily exchanged basic human need – shelter, not primarily an asset with massive financial inertia to change)
But that’s expensive, in £, time, administration, politics.
Um, yes, that’s how evidence works.
Perhaps you meant they’re not *proof*, which would be true. But they are the best evidence we have.
@mdavidford
Good point, I meant and should have said conclusive, or proof as you suggest. 🙂
Independent variable = speed limit
Dependent variable = casualties
Control variable = roads where no change made
The figures prove that when we change the independent variable by lowering the speed limit then the dependent variable, casualties, falls in comparison to the control variable of roads where no change to the speed limit has been made. That is evidence.
@Rendel Harris
If there are no other variables.
Be mindful of those that present such simple arguments as evidence.
If there is no consideration for frequency of car usage, frequency of bicycle usage etc. then the stats can only be said “appear to suggest”, as correctly stated by the road.cc author.
These stats include the era of the post Iran fuel price spike of 40%, where car usage has fallen. Which could skew the numbers unless the stats are adjust by number of journeys, road users etc.
For what it’s worth I just want to point out potential abuse of stats. Not an opinion on 20mph zones or a desire to kill cyclists as some might contort it to be.
@Gd29 All of those are factors that could be influential however you do not address the fact that the dependent variable has shown a fall against the control variable. If the control variable had fallen at the same rate as the dependent variable certainly that would require investigation, but it hasn’t. If we regard 20 mph limits in Wales as an experiment, we can say that the conditions of the experiment were such that only one change was made that didn’t affect all of the roads in the experiment and a significant result has ensued from that change. That is evidence.
I see nothing to indicate that you want to kill all cyclists, however.
@Gd29 You accept going too fast is an issue but don’t accept that reducing speed and falling accident/injury rates are related by anything other than correlation? In which case, how do you begin at your own starting point? Either speed is an issue or it isn’t. And if drivers won’t drive to the conditions, and we all agree speed is an issue, then forcing them to drive at lower speed is a perfectly reasonable solution. There is no harm in also educating drivers in better habits, but that is a much more difficult and timely process. So how many deaths and serious injuries are acceptable in a world where we keep 30mph as the standard limit on relevant roads whilst we wait for drivers to catch up in a utopian world.
I do not disagree about most of the other solutions you put forward. Some of them, however, are even less likely to occur within the next generation or two than materially improved driving behaviours.
@Gd29 exactly what variables / controlled experiments would suffice to convince you, and then how would you believe that would still hold in “real world” situations?
We all know there are excellent ways to stop motoring-related KSIs – one being stop humans operating vehicles) but that isn’t normally seen as realistic.
Then there’s another “remove the hazard” approach eg. completely isolate motorists from vulnerable road users (and “footways” alone might not be sufficient, as more people are killed there most years by drivers than eg. by cyclists in total, although that fact doesn’t say much ). Though that by itself doesn’t address motorists killing themselves / other motorists.
Currently the UK – like many countries – is mostly doing a “remove the vulnerable road users” approach, partly by design but mostly indirectly by prioritising convenience for motoring (including journey time, which is seen to involve higher speeds).
@Gd29 it’s good to “point out potential abuse of stats – and for “keeping it concise for the public” no doubt some measure of “lies, damn lies and statistics” applies.
But unless you have your own analysis / evidence, this may sound like a “just asking questions!” rhetorical tactic.
@Gd29 “Speed limits are an ineffective tool when the problem is drivers not driving in an appropriate manner for the conditions and environment. Those drivers weren’t following the speed limit and/or paying sufficient attention anyway!”
That is a question, not an assertion. What are the relative numbers of drivers who are eg. “don’t care” criminals who may never have taken a test or are happy to take life, how many are “criminally negligent” eg. may have a licence and not intend to break the law regardless, but just aren’t fussed about basic safe driving (maybe regularly drive while tired, never slow down when conditions require), how many are just normal humans (who occasionally make unfathomable mistakes)? What are the relative rates of collisions? What are the casualty outcomes?
What effect does simply changing a number on a sign have? If any, why? (There’s the psychological “anchoring effect”, if 70% of people *are* below the limit does that slow down others, do the “don’t cares” maybe just overtake dangerously (but would they anyway? ) ) etc.
Reduced speed is a mitigation in that it leads to better outcomes in collisions, that’s a known I believe. IIRC there’s also agreement that it can mean that fewer collisions happen as there’s more time to react.
“If that is the problem and you don’t deal with it, but you want to keep reducing injuries/fatalities, you have to keep dropping speed limits until you prohibit road use entirely”
“If” (plus as noted other safety measures are available)
@Gd29 I was waiting for the classic “why don’t we just ban driving completely by reducing the speed limit to 5mph or 0!” argument and here it is.
The statistics are very clear on the benefits of reducing the speed limit from 30-20mph. It doesn’t remove bad driving but it mitigates the damage. It also makes fuck all difference to journey times.
Everything with driving is risk vs reward. Sure, dropping it to 10mph would save a lot more people but it would make travel by car ridiculous and have minimal benefit vs 20mph. The reward wouldn’t be worth the cost.
@mctrials23 yeah … but that’s when we look at places like car parks and conclude that in fact 10 or even 5mph does in fact make a lot of sense!
Having experienced this in Anglesey and most of it is unnecessary, it does give the driver a big lesson in ‘looking extra’, as travelling slower does that !