Article Text

Download PDFPDF

Twenty miles per hour speed limits: a sustainable solution to public health problems in Wales
  1. Sarah J Jones,
  2. Huw Brunt
  1. Public Health Wales, Cardiff, UK
  1. Correspondence to Dr Sarah Jones; sarah.jones27{at}


Background Prevention, rather than treatment, is the key to longer healthier lives. Identifying interventions that will impact positively on road traffic injuries, air quality and encourage active travel is a significant public health challenge. This paper aimed to explore whether 20 mph limits could be useful in achieving this.

Methods Research evidence was reviewed to identify the effect of 20 mph zones and limits on health and well-being. The evidence was then used to estimate the effect of a change to a 20 mph limit on road traffic casualties and air pollution. It was then mapped against the seven goals of the Well-being of Future Generations Act (2015).

Results If all current 30 mph limit roads in Wales became 20 mph limits, it is estimated that 6–10 lives would be saved and 1200–2000 casualties avoided each year, at a value of prevention of £58M–£94M. In terms of air pollution, deaths attributed to nitrogen dioxide (NO2) may increase by 63, and years of life lost by 753. However, deaths attributed to particulates (PM2.5) may decrease by 117 and years of life lost by 1400. Evidence review suggests benefits in terms of road traffic casualties, air quality, active travel, noise pollution, greater social inclusion, greater community cohesion and local business viability.

Conclusions Road traffic injuries, air pollution and obesity are an inter-related, interdependent triad. The challenge facing public health today is identifying robust interventions that will have positive effects on all three as a minimum; default 20 mph limits is the solution to increasing public health problems in Wales.


Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


Prevention, rather than treatment, is the key to healthier, happier and longer lives. At the population level, collaborative action is needed across public bodies and communities to protect and improve health. This is recognised by the Well-being of Future Generations (Wales) Act 2015 (WFGA),1 which calls for collaboration to achieve sustainable health and well-being improvements.

The act lists seven well-being goals, one of which is ‘a healthier Wales’, and requires a broad, inclusive view to be taken to tackle identified problems (figure 1). It encourages the proposal and implementation of public health interventions that contribute to as many of these goals as possible to maximise health gains. However, identifying interventions with proven benefits across a range of health effects and few, if any, negative effects is a significant challenge.

Figure 1

The Well-being of Future Generations Act goals.

Adopting 20 mph as the default speed limit has the potential to be such an intervention. Currently, 20 mph zones are commonly in use over short distances outside schools. Zones include traffic calming which is intended to force vehicles to slow down. Limits, however, are usually only noted by signs posted on poles at the road side and roundels painted onto the road surface. A default 20 mph limit has been argued to just ‘make sense’.2 Road safety charity Brake is actively campaigning for a 20 mph default, Royal Society for the Prevention of Accidents (RoSPA) for wider use of 20 mph zones. National Institute for Health and Care Excellence (NICE) and the British Medical Association (BMA) have recommended that city-wide or town-wide 20 mph limits should be introduced.3 ,4 The Faculty of Public Health believes that 20 mph zones and limits are important to mitigate the health impacts of cars.5 Royal College of Paediatrics and Child Health (RCPCH) Wales are also calling for 20 mph limits to encourage children in Wales to be healthy and physically active.6

Much of the focus to date has been on road safety, but the suggested benefits extend beyond this to improved air quality,2 ,7–9 increased active travel,2 ,10 narrowing of inequalities,2 ,4 ,11 greater social inclusion,2 ,11 reduced noise pollution9 ,11 ,12 and greater community cohesion,2 ,11 including viability of local businesses2 (table 17 ,9 ,11–18).

Table 1

Effect of 20 mph zones and limits on health and well-being

24 European OECD countries (EU24) estimates that 50% of air pollution deaths are due to road transport12 and there is evidence that changes in driver behaviour linked with lower speed limits could reduce emissions.9 For example, nitrogen oxides (NOx) emissions under hard acceleration are two to four times higher than those of constant speed.19 In Sweden, 20 mph limits have been associated with lower fuel use because of less starting and stopping, compared with 30 mph limits.20 Lowered speed limits, rather than speed humps, also have a positive effect on emissions.20 While not all of the available evidence suggests that 20 mph limits decrease all forms of air pollution, it is generally accepted that they do not increase air pollution relative to 30 mph limits.21

In the UK, high vehicle speeds are often cited as preventing people from walking or cycling.10 Reducing speeds should therefore encourage walking and cycling. Public Health England has suggested that default 20 mph limits may be the most effective evidence-based approach available to increase exercise levels,22 but no significant changes and even decreased walking and cycling have been reported in 20 mph zones.23

Real or perceived road danger can also have an opportunity cost in terms of walking or cycling, meeting friends and family and other recreation, as well as the freedom to develop independence.4 Car domination is suggested to have led to the closure of local shops,24 lower speeds are believed to make local shops and services more commercially viable11 and people who travel to shops on foot spend more than those who do not.25

The most deprived areas have more types of vehicle travelling at a greater range of speeds26 and a 20 mph limit default could disproportionally benefit those previously at greatest injury risk.2 ,4 ,11 Effects on crashes, emissions, noise, health or community severance could therefore be greatest in the most socially excluded areas.2 ,4 ,11

The appetite for 20 mph zones and limits is high among charities, the public sector and the general public; 77% of survey respondents favoured residential 20 mph zones, rising to 89% support around schools.27 Regulatory changes have made it easier for local authorities (LAs) to introduce 20 mph zones. But, too local an approach also has limitations.

The WFGA means that acting on this evidence base in Wales is more appropriate than ever. In addition, the Draft Wales Bill28 proposes giving legislative competence for speed limits to Wales, making the implementation of a national 20 mph default possible. However, research evidence, especially from other parts of the world, is rarely enough to convince policymakers of the need to act; the potential public health benefits need to be quantified wherever possible, and the relevance to policy objectives demonstrated.

This paper makes the case for 20 mph limits in Wales by estimating potential health effects of a change and applying WFGA in a clear, simple way. Other methods may be more accurate, but at the cost of losing clarity. This ‘approximately right’ perspective29 supports decision makers by quantifying public health effects of interventions.


Road traffic casualty effects

Police crash data (STATS19) for 2011–2013 were used to identify all crash casualties injured on roads in Wales with a 30 mph limit. The number of fatal, serious and slight casualties was calculated for each year and the effect of 20 mph limits estimated by reducing the actual numbers by 42%,13 25%6–15 and 4%; the effect needed to save just one life (table 2). Forty-two per cent was the finding of the most recent study; however, it is a study of 20 mph zones and it is not known how this would translate to countrywide 20 mph limits. The effect size of 25% was also found in a study of 20 mph zones.15–16 However, given the age of these studies, the applicability is again questionable. Therefore, a final analysis was also carried out based on the effect size needed to save just one life (4%).

Table 2

Worked example of crash casualty effects calculation (2011 data)

The value of prevention was calculated using UK Department for Transport (DfT) figures for 2012.30 Data for each of the study years were aggregated and an estimate of the annual average casualty and cost savings associated with road traffic casualties avoided with 20 mph limits was calculated.

Air pollution effects

Estimates of the annual average concentrations of NOx, nitrogen dioxide (NO2), particulate matter of less than 10 μm diameter (PM10) and Particulate Matter of less than 2.5 μm diameter of less than 2.5 μm diameter (PM2.5) for 2011–2013, at residential dwellings, by LA, were obtained from Welsh Government31 (table 3, row A). There are 22 LAs covering the 3.2 million people in Wales and they are responsible for delivering local services and setting local speed limits.

Table 3

Worked example of air pollution effects calculation

The transport-related component was then estimated as 33% of the total for NOx.32 For PM10, 30% was estimated based on a range of effect sizes available (see online supplementary 1).

supplementary data

Here, modelled air pollution data were used because they more accurately reflect area concentrations and population exposures.33 These data are validated annually against measured data, but still may not correspond to actual individual exposure, which is influenced by factors such as activity patterns and time spent indoors and outdoors.34

For each LA, the total road length, 30 mph road length and proportion that is 30 mph was calculated using ArcMap 10.3.1 and OS MasterMap. These outputs were then used to estimate the air pollution concentration associated with 30 mph roads (row D).

The effect of changing to a 20 mph default was then estimated based on the relative proportions of diesel and petrol cars on the road in 2014 (1 diesel:2 petrol,35 row E) and data from a study by Imperial College, London based on the Euro IV emission standard and 1.4–2.0-l vehicles.21 From this, it was estimated that NOx emissions would increase 7.6%, while PM10 would decrease 24.9%. Here, the measured data were most appropriate.

Estimates for NOx and PM10 were then converted to NO2 and PM2.5 using the Department for the Environment, Food and Rural Affairs (DEFRA) NOx to NO2 calculator (v4.1)36 and, for PM, a conversion factor of 0.65.37

Correlation coefficients for all-cause mortality were then applied (1.062 for PM2.5 and 1.025 for NO2;38 row G) and the relative risk (row H39) and attributable fraction (row I) calculated. Other coefficients for NO2 were considered (see online supplementary 2) but the Committee on the Medical Effects of Air Pollutants (COMEAP) value was deemed most appropriate because they are the advisory group on the subject in the UK.

supplementary data

All-cause non-accidental mortality data for over 30-year-olds, as recommended for these types of analyses,40 were obtained from NHS Wales Informatics Service for each LA for 2011–2013 and used to calculate average numbers of deaths per year (row J). Numbers of attributable deaths (row K) and years of life lost (YLL, row L41) were then calculated by LA and summed for the whole of Wales.

This gave an estimate of attributable deaths and YLL once 20 mph limits have been introduced. The effect is therefore the current effect associated with 30 mph roads minus the estimated effect of a switch to 20 mph limits. The current effect was calculated using the 2012 estimated average PM2.5 and NO2 concentrations for the LA.36 This value was entered into the table at line A and steps B, C and D repeated. The value obtained at D was then entered into line H, and the steps through to L repeated. The savings were then estimated.

Effects were monetised in life years. Rabl and De Nazelle29 took a similar approach, using a value for prevention figure for a crash fatality (€1.6M) and a life years value for air pollution (€43 801), arguing that this was appropriate because road traffic crashes tend to affect the young, leading to losses of half a life span or more, while air pollution reduces life expectancy by months. The value of €43 801 for 201029 was converted to £40 484, using a currency conversion tool and the developer's recommendations from a linked project,42 and adjusted for inflation to £43 958.43

Policy impacts

The available evidence about the effects of 20 mph limits was considered in the context of the seven well-being goals of the WFGA.


Road traffic casualty effects

In Wales, between 1 January 2011 and 31 December 2013, 14 639 people were killed or injured on 30 mph limit roads. Reducing all current 30 mph limit roads to 20 mph limits could prevent 6–10 deaths and 1203–1978 injuries per year at a value of prevention of £58M–£94M (table 4).

Table 4

Estimated road traffic casualty savings following the introduction of 20 mph speed limits

Air pollution effects

Switching from 30 mph limits to 20 mph limits may lead to a slight increase in deaths and YLL associated with NO2, but a greater reduction in deaths and YLL for PM2.5 (table 5).

Table 5

Estimated changes in air pollution related deaths and YLL following the introduction of 20 mph limits

Policy impacts

Each of the goals in the WFGA has a detailed description associated with it (table 6); these provide the detail against which the proposed speed limit change can be evaluated.

Table 6

Contribution of changes in default speed limits from 30 to 20 mph to achieving the goals of the WFGA


A change to a default 20 mph speed limit has the potential to have significant positive effects on the health of the people of Wales and deliver against all seven WFGA goals. We have attempted to estimate some health effects, but it is likely that there will be many more cumulative positives that are greater than individual effects, and are beyond our ability to estimate them here. Even so, the road casualty savings alone are sufficient to make the case for change. Interpreting the air pollution effects is more difficult, but the suggestion is towards positive health effects. In addition, while the direct air pollution effects are equivocal, indirectly, there will be positive effects as more people are encouraged to participate in active travel options, thereby having a longer term benefit that is impossible to calculate.

This paper adds to the existing evidence base by attempting to bring together estimates of the effect of default 20 mph limits on multiple public health priorities in one paper. The importance of this is that ‘putting numbers on the harmful effects of air pollution allows policymakers to compare the costs of action with the benefits that will follow’.24 It is about making the case for a national change, rather than an ad hoc localised approach, as currently seen, which may cause confusion and be ineffective. In terms of air pollution, a localised approach is likely to be detrimental because emissions are affected by driving style, in particular acceleration and deceleration.7 ,9 ,44 To bring about changes in driving style will need much wider 20 mph limit use. In addition, a universal 20 mph limit avoids displacing motorised traffic onto other local roads, increasing risks elsewhere and creating inequalities.

Much of the existing evidence is based on changing a small number of roads in ‘high-risk’ areas, usually based solely on injury risk, for example, around schools, from 30 mph limits to 20 mph limits. The approach that we are proposing is more widespread; a blanket reduction of all 30 mph routes in Wales to 20 mph limits, whether urban or rural. Theoretically, this is consistent with Rose's prevention strategy45 and would suggest that focussing speed reduction on a few high-risk areas, schools, has a smaller effect than reducing speeds across a much wider area. More widespread use of 20 mph limits would also have benefit in terms of the simplicity of such a change; drivers are currently taught that if there are no repeater signs then the road is a 30 mph limit. However, 20 mph limits do currently carry repeaters. By applying a blanket change, some of the existing signs can be removed, a DfT goal and urban driving can be simplified. Policing and enforcement would also be simplified. This would all reduce the costs of a 20 mph change. This also means that the effect will be greater than we predict because the data on which these analyses are based relate to small traffic-calmed areas.

Do we need a precise and detailed estimate of effect? We would argue not. This is a broad analysis to illustrate the potential of 20 mph limits across a range of health effects in the context of WFGA. Given that we do not know how these interact, we believe that a simple approach is far more useful than trying to add further layers of complexity to an already difficult analysis. In addition, while we have in many ways taken a best case view of the data, there is little evidence to suggest that a nationwide 20 mph limit will cause harm. When reviewing the existing literature, ‘no effect’ on air pollution was the worst outcome that we found;44 however, this was based on research into short 20 mph zones covering a small area, often unenforced and with little potential for effect on driver behaviour, the main factor in reducing emissions. According to the literature, therefore, changing all current 30 mph limits in Wales to 20 mph may have no effect on crashes, air pollution, physical activity levels, inequalities, play and the many other factors that the literature suggests are influenced by reducing speeds. But, based on the evidence, such an outcome is highly unlikely. We believe, based on the evidence generated here, that the potential benefits are significant.


These estimates are based on injuries and air pollution and treat them equally, but the injuries’ analysis is based on actual casualties and fatalities whereas the estimates relating to air pollution are ‘statistical deaths’.24 Identifying robust data with which to analyse wider health impacts was difficult; for example, the contribution of noise pollution to stress and anxiety and how reducing motorised traffic noise could reduce these and associated physical health effects.

Analyses are based on research into 20 mph zones; the cumulative effect of nationwide limits is likely to be greater than is estimated here.

Analyses take no account of actual changes in vehicle speeds. It is understood reductions in actual speeds would be unlikely to reach the 10 mph change in the limit. For this reason, we have tried to use data from 20 mph zones, rather than modelled effects.

With a national analysis, it is also impossible to model the effect on travel times. But, evidence suggests that speed limit changes affect travel time far less than people imagine.9

Limitations—road traffic casualties

These analyses are based only on overall casualties and not adjusted for age and road user type. However, STATS19 is known to undercount pedestrian and cyclist casualties,46 so a more detailed analysis would not necessarily improve understanding.

The analysis has not accounted for potential increases in cycling/pedestrian casualties that may occur because there are more of them. However, the critical mass effect suggests that more cyclists and pedestrians will reduce the risk to all.47 In addition, collisions that do occur will be of lower speed and recent estimates suggest that at 20 mph there is a 1.5% risk of death, compared with 5% at 30 mph.48

Limitations—air pollution

The effects on air pollution were based on estimates generated for cars only; there are no estimates for changes to emissions by heavy goods vehicle (HGVs), buses or other light passenger vehicles. It has been suggested that around 40% of PM10, 50% of NOx and 50% of PM2.5 emissions are by HGVs, buses, coaches and light commercial vehicles.49

We assume a linear relationship between the proportion of road that is 30 mph and the contribution of those roads to the overall burden of air pollution. In reality, 30 mph roads are likely to contribute more air pollution than is suggested by their length because of the range of vehicles, the numbers of vehicles and the stop start nature of the motorised traffic on them. Therefore, this approach underestimates the effect of a switch to 20 mph on air pollution.

The basis for these estimates is place of residence. In reality, exposure occurs in many different places, including in the car. It has been suggested that car occupants are exposed to three times more emissions than cyclists and pedestrians because of their seat in the pollution tunnel in the centre of the road.24

These estimates are based on short-term air pollution changes in small 20 mph zones. They do not account for long-term changes as people shift to other, more sustainable, modes of transport and active travel. They therefore underestimate the true effect.

Currently, many 20 mph zones use speed humps and cushions and although these reduce speed and therefore crash risk, they help to increase emissions.9 With a default 20 mph, it would, arguably, be possible to reduce the use of traffic calming, which would also then further improve air quality, beyond what we can estimate here. Speed humps and cushions also increase road maintenance and repair costs, not just because of the costs linked to the features themselves, but also because of the damage to the road surface around the feature.11

Two pollutants have been considered independently and mortality burden estimates presented separately; summing them would overestimate attributable mortality38 and there is believed to be a 30% overlap of mortality burden attributed to PM2.5 and NO2.41 There is not yet an agreed method for dealing with this overlap when calculating the burden.

While it would have been useful to calculate a range of estimates of effect for air pollution in the same way as with road traffic casualties, there are insufficient data available on the effects of 20 mph zones/limits on air pollution, compared with 30 mph, to make such analyses feasible.

Limitations—current length of 30 mph road

There are already 20 mph zones around schools in Wales and it has been estimated that 20% of urban streets in Great Britain (GB) are already 20 mph,50 meaning the proportion of road length that is currently 30 mph is overestimated, as are the effects of the change. However, current 20 mph zones make up only a short length of overall road and enforcement and observation of this is poor. Therefore, while these data may overestimate effect in places, current effects are likely to be small.

These analyses are based on a change in all 30 mph roads, but transport authorities believe that there are some major routes that it would be inappropriate to change to 20 mph. These are likely to be routes that carry the most trafficked and widest range of vehicles, meaning that 30 mph is retained in the most deprived areas, potentially creating greater inequalities.

Limitations—value of prevention

The estimates of the value of prevention are presented, but there are no estimates of the cost of implementation. Peters and Anderson51 found that 20 mph zones were cost-effective in high-casualty areas only, but, their analysis covered only injuries.

Wider context

The costs of air pollution are expected to rise, as is the volume of motorised traffic on our roads. Identifying robust, low-cost interventions to address this is difficult, but 20 mph limits have the potential to be that intervention, not least because of reductions in indirect costs, such as road network costs, highlighted here. It has been suggested that the costs of air pollution are so high that greater investment in tackling the problem will still yield a positive benefit.24

Of existing interventions, 95.8% of air quality management area breaches are thought to be due to transport emissions.49 It has been suggested that continued focus on technical controls to abate vehicle emissions offers less benefit for public health than measures that increase active travel and public transport;24 a 20 mph default limit would address this.

The risks to children are key here, not least because of the harms that they suffer because of speeding traffic; road traffic injury risk deters parents from allowing them to play outside and walk or cycle to or from places, risking obesity, reducing activity levels and reducing the opportunities for socialisation. If these children are allowed to play, walk or cycle near roads, because they are at a similar height to motor vehicle emissions and have a faster breathing rate than adults, they suffer double jeopardy; more exposed and more vulnerable.24 In addition, height and smaller stature means that if these children are hit by vehicles, the head is a primary point of contact and the risk of serious injury is much higher than for adults. So, parents decide that it is ‘safer’ to transport their children by car but air pollution levels inside vehicles are much higher than those outside because fans and air conditioning units drawing fumes from exhausts directly into the vehicle.24 Air pollution can also affect the growth of the unborn baby and may be linked to premature birth.24

This paper focuses on the benefits to future generations. However, 20 mph limits will create a more supportive environment for vulnerable older road users. Older pedestrians struggle to cross roads when motorised traffic is fast moving and older drivers have difficulty joining fast-moving traffic.


The evidence presented here suggests a national 20 mph limit would substantially reduce road traffic casualties and, at worst, would not lead to a direct change in air pollution. Indirectly, if people are encouraged to switch to active travel, there are likely to be greater reductions in air pollution. The health and cost savings of the approach are likely to be substantial and although the costs of implementation are not accounted for here, they are likely to be far lower than the benefits. It has been suggested that a Wales wide 20 mph default would be eight times cheaper than a town-by-town implementation.52

The DfT has commissioned research into the effectiveness of 20 mph limits and the results are due in 2017.53 The study will cover ‘speed, collisions, injury severity, mode shift, quality of life, community, economic public health benefits and air quality’. However, regardless of the conclusions of this report, the draft Wales Bill 2016–2017 proposes giving Wales powers to set its own speed limits and providing the opportunity to realise significant public health benefits.

Scotland already has these powers and advocates there have been calling for a 20 mph default since March 2016 when the law was passed. In Ireland, the Road Safety Authority is calling for much wider use of 20 mph to reduce injuries and increase tourism.54

We have analysed injury and air pollution effects. We will aim to build on this in the future to analyse the impacts on health associated with, for example, changes in noise and physical activity levels.

Caution and expectation management are still needed. The effect on road traffic injuries should be seen relatively quickly once 20 mph limits are introduced. Increases in active travel would follow once people felt that roads were safer. For air pollution, although changes in emissions should begin relatively quickly, there are likely to be much slower reductions in ambient concentrations10 and the health impacts may take many years to be realised.

Some may also argue that a switch to electric vehicles will deal with many of these concerns. However, electric vehicles can still kill and injure, tyre and brake wear produces particulate air pollution and vehicle movement disturbs road dust. It is estimated that particulate production is 50% emissions, 50% degradation, and it is known that higher traffic speeds create more air turbulence that spread the particulates further.55 Logically, lower speeds will create less friction, reducing tyre wear, require less braking, reducing brake wear and will spread the particulates that are produced less far.

Road traffic injuries, air pollution and obesity are an inter-related, interdependent triad. The challenge facing public health today is identifying robust interventions that will have positive effects on all three as a minimum; default 20 mph limits is the solution to increasing public health problems in Wales.

What is already known on this subject

  • Use of 20 mph zones can have positive effects on a wide range of health outcomes.

What this study adds

  • An estimate of the population level benefit across more than one health outcome.

  • An assessment of the relevance to current policy.

  • It concludes that a shift to a 20 mph limit default could have significant benefits for road traffic crash casualties, air pollution and active travel. It could be an extremely important public health intervention.



  • Contributors SJ and HB developed the idea for the paper, the methods and drafted and edited the paper.

  • Competing interests None declared.

  • Provenance and peer review Not commissioned; externally peer reviewed.