Response to Department for Transport Consultation on Revised Speed Limit Setting Circular

September 2012


The ABD was formed in 1992 to campaign for a better deal for Britain’s motorists. It is a voluntary organisation funded by subscriptions and donations from members and supporters. It currently has around 1,000 members and a further 3,000 supporters on social media sites. The ABD receives no funds from public bodies or private-sector businesses, so is truly independent. It is a member of the Parliamentary Advisory Council for Transport Safety and the National Council of Voluntary Organisations.

The ABD’s national committee has agreed the response to this consultation. The response is in line with the policies outlined on the ABD’s website. Individuals who join or support the ABD are assumed to support its policies in general, so it is not necessary (nor would it be practicable) to seek the views of the entire membership in preparing responses to consultations of this type.

General Comments

Before answering the specific questions in the consultation document, the ABD wishes to comment on the principles behind the draft speed limit setting circular. In the ‘key points’ section within the introduction, the first paragraph is as follows:
Speed limits should be evidence-led and self-explaining and seek to reinforce people’s assessment of what is a safe speed to travel. They should encourage self-compliance. Speed limits should be seen by drivers as the maximum rather than a target speed.
The ABD agrees wholeheartedly with the sentiments expressed in that paragraph. It is the way in which the ABD has promoted the correct use of speed limits since its formation twenty years ago. Sadly, the basis of the draft revised circular, which is the same as that of its 2006 predecessor, will fail to deliver those desirable goals.

Assuming that the distribution of speeds follows approximately a normal curve, if speed limits are set at the mean speed, half of all drivers will regard the limit as being too low. The limit will not, therefore, reinforce people’s assessment of what is a safe speed to travel but will conflict with that assessment. The speed limit will not encourage self-compliance either; rather it will lead to a high level of non-compliance.

Evidence previously submitted by the ABD to the Department for Transport shows that the drivers with the least accident involvement are those who travel in the 80th to 90th percentile range of the speed distribution. Consequently, setting speed limits at the mean speed needlessly criminalizes the safest drivers if they continue to use their judgement to travel at a safe speed for the conditions.

Paragraph 13 of the draft circular repeats the claim that a reduction in average speeds of 1 mph reduces accident frequency by 5 per cent. The studies from which this and similar claims are drawn have been shown by the ABD to be statistically flawed, based as they are on comparing casualty rates on roads of allegedly similar characteristics but different average speeds. Those differences in average speed are due to different traffic flows, but the relationship between speed and flow is non-linear and this is not adequately compensated for in the studies. Indeed, the Transport Research Laboratory has specifically recommended that roads with markedly different traffic flows should not be studied together (PPR026, paragraph 4.15).

In any case, accidents are not caused by high average speeds but by speeds that are too high at specific locations under particular circumstances. Slowing drivers down needlessly where higher speeds are safe does not guarantee that they will be travelling at an appropriate speed at the hazardous locations. Instead, forcing drivers to travel below the natural speed of a road will lead to lack of concentration and greater liability to distraction, so they will be less likely to spot hazards developing ahead.

Paragraph 14 of the draft circular seeks to manipulate the reported contributory factors to road accidents in order to increase the significance of speed in accident causation, in an attempt to justify driving down speed limits. Exceeding an arbitrary numerical speed cannot, of itself, cause an accident. Inappropriate speed for the prevailing conditions, whether above or below the speed limit, can certainly be a contributory factor. Except where criminal activity is involved, however (e.g. drink or drug driving, unlicensed, etc), inappropriate speed is usually the result of a failure by the driver to observe, or correctly respond to, a potentially hazardous situation. This may be the result of inexperience or, in the majority of cases, lack of attention to the driving task. The latter is the root cause of the vast majority of accidents.

Speed limits have a role to play in road safety and, when set correctly, they provide a positive contribution in the following ways: If set too low, however, they are incapable of performing those functions and can instead have a negative effect on safety. The ABD has always advocated the use of the 85th percentile speed as the basis of setting speed limits, as there is evidence stretching back several decades that this method produces: The trend towards reducing speed limits below the 85th percentile, which began in the 1990s and was ‘legitimised’ in the 2006 circular, has eroded those benefits. The move to using the mean speed for speed limit setting was based on flawed research and false beliefs about the purpose of speed limits and their effectiveness in reducing both actual speeds and accidents. The result has been an increasing lack of respect for speed limits, leading to greater enforcement, often applied inflexibly by speed cameras, to force drivers to conform to unreasonable and unnecessary limits. This in turn creates resentment, as large numbers of drivers have been prosecuted for mainly technical offences where no danger has been caused.

This downward spiral needs to be halted and reversed. The only way this can be done is to revert to the 85th percentile speed for setting limits and to make its use mandatory on traffic authorities. The draft circular should be rewritten, therefore, based on that principle.

Response to Specific Consultation Questions

Q.1 Do you agree that the advice about introducing 20 mph zones and limits provides useful guidance to traffic authorities considering speed management in urban areas?
The advice on 20 mph zones and limits contains some misleading or inaccurate statements. Paragraph 70 repeats the claim that there is a reduction in accident frequency if average speeds are lowered, but this is again based on a flawed study. While lower impact speeds with pedestrians will, other things being equal, lead to less serious injury, road safety policy should not be based on hitting people more slowly but on avoiding hitting them altogether. Also, merely changing the speed limit does not guarantee any reduction in actual speeds.

Even if driving speeds do reduce as the result of a lower limit, there is no certainty that impact speeds will fall. A driver travelling at lower speed may be less alert and not react quickly enough if, say, a pedestrian walks into the road. If the speed limit is below the natural speed at which a driver would choose to travel, it will require more frequent glances at the speedometer to stay within the limit, thus further increasing the chances of a potentially hazardous situation not being spotted in time.

Paragraph 84 claims that Portsmouth’s 20 mph limits have led to some reductions in casualties, but a detailed analysis of the figures, once national trends and changes in traffic flow have been allowed for, show that injuries to some classes of road user have increased significantly. Other 20 mph schemes in Bristol, Oxford, Warrington and St Albans have also shown increases in casualties. It seems likely that the introduction of 20 mph limits creates a perception of safety on the part of some road users, so they become complacent and take less care.

Paragraph 77 states that 20 mph zones are very effective at reducing collisions and injuries, but this is largely because the traffic calming installed within them leads to significant diversion of traffic, which is often one of the aims of the schemes. This traffic will be redistributed to the main road network but, because the numbers of vehicles involved will be small compared with the traffic flow already on those main roads, any resulting increase in accidents is likely to be within year-to-year random variation and will thus be undetectable.

In paragraph 73 it is stated that a 20 mph zone or speed limit can be regarded as self-enforcing if it results in a mean traffic speed compliant with the limit. This would mean, however, that half of all drivers would still be travelling above the speed limit, so this can hardly be considered ‘compliant’. Presumably the assumption is that a mean speed of 20 mph is likely to equate to an 85th percentile speed of around 24 mph (the ACPO enforcement guideline), so no enforcement would be considered necessary. As noted in the general comments above, imposing a statutory speed limit with the expectation that it will be broken by half of all drivers does not seem a satisfactory approach to the rule of law.

Paragraph 71 claims that 20 mph schemes “include quality of life and community benefits, and encouragement of healthier and more sustainable transport modes such as walking and cycling”. This is nothing but wishful thinking, as the effect of signed-only speed limits on actual speeds will be minimal. Traffic-calmed 20 mph zones will certainly see a reduction in traffic volume and speeds, but that traffic will simply relocate to other roads. The claim that “driving more slowly at a steady pace will save fuel and reduce pollution” is risible. Most modern cars use less fuel at a steady 30 mph than at 20 mph. In 20 mph zones, the traffic calming features cause drivers to constantly accelerate and decelerate, wasting fuel. It must also be remembered that catalytic converters only work properly when they are hot – in stationary or slow-moving traffic they cool down and fail to remove pollutants effectively.

Although the ABD does not oppose all 20 mph speed limits in principle, they should only be applied where the 85th percentile speed is already 20 mph or below, such as in narrow residential streets with on-street parking. Of course, if speeds are already that low, reducing the speed limit from 30 mph will not provide any benefit and could actually lead to an increase in speeds, if they were previously below 20 mph, as drivers would be reminded that they are legally entitled to drive up to that speed. (In a street-lit urban area, the absence of repeater signs means drivers are not reminded that they are legally entitled to drive at 30 mph, so they vary their speed according to the conditions without thinking about the speed limit.) The experience of Portsmouth and elsewhere suggests that road users may take less care in a 20 mph limit, so the anticipated improvement in road safety may not be realised. There is certainly no evidence that the widespread use of 20 mph limits is likely to be beneficial, so a cautious approach should be taken.

Q.2 Do you agree that traffic authorities should be able to consider the implementation of 20 mph limits over a number of roads where mean speeds at or below 24 mph are already achieved?
The ABD considers it unacceptable to introduce a speed limit at the existing mean traffic speed, let alone below it. If the mean speed were as high as 24 mph, most drivers would be exceeding 20 mph, so a very high level of non-compliance would result if a 20 mph limit were introduced. Not only would this devalue speed limits generally, it would increase risk to vulnerable road users, as most people have an unrealistic expectation of the ability of reduced speed limits to produce lower driven speeds. They would be lulled into a false sense of security and take less care, as the negative results from Portsmouth and elsewhere have shown.

Q.3 Do you agree that the recommendation to use the technical assessment tool should be withdrawn?
It is the stated intention to replace the technical assessment tool in the 2006 circular with a new Speed Limit Appraisal Tool, so it is logical that the previous tool should be withdrawn. The ABD assumes the revised circular will not come into force until the new appraisal tool has been developed and subject to consultation. Only then will it be possible to assess whether the new tool will be an improvement on the old one.

Q.4 Do you agree that compliance with air quality limits could be a factor in the choice of speed made by local traffic authorities?
Speed limits have always been introduced on the basis of claimed benefits to road safety and it would be intolerable for drivers to face fines and penalty points for exceeding limits introduced for any other purpose. The majority of locations with air quality problems are in urban areas, where it is congestion and low speeds that are the cause. Some types of vehicle contribute disproportionately to the emission of certain pollutants, with buses and heavy goods vehicles often the main cause of poor air quality in city centres, where speed limits are already low in any case.

Low speed limits imposed for air quality reasons on major roads such as motorways, if they happened to pass close to a built-up area, would lead to massive non-compliance. Heavy goods vehicles, which make up a significant proportion of traffic on such roads, would hardly be slowed at all, so their contribution to air quality would be unaffected. At periods of low traffic flow, such as the middle of the night, the air quality problem would recede so the limit would be pointless.

As vehicles with cleaner engines replace older ones, air quality should improve even if traffic volumes rise. The answer to air quality issues is to ensure road capacity is sufficient to allow traffic to flow freely, and to tackle congestion in urban areas by removing underused bus lanes, retiming traffic signals, and so on. Speed limits must not be used for reasons other than road safety.

Q.5 Do you have other comments about the drafting of the revised circular?
In addition to the general comments set out at the beginning of this response, the ABD has the following points to make on specific paragraphs in the revised circular:

Paragraph 22 suggests that, among the factors to be taken into account in setting local speed limits, the desires of local residents for a lower limit should be considered. Most people think, wrongly, that lower speed limits lead to lower speeds and fewer accidents. In reality, speed limits set too low lead to a high level of non-compliance and a greater spread of speeds than when they are set correctly at the 85th percentile level. When set at this level, the fact that most drivers observe the speed limit puts peer pressure on the reckless or inexperienced minority to conform. There is no such pressure when the limit is too low, with the result that, paradoxically, there may be an increase in the highest speeds compared with a higher, more respected speed limit. It was because of some authorities pandering to the demands of uninformed residents for lower speed limits that we have arrived at the unsatisfactory situation we see today. Speed limits must be set according to scientific principles, not on the basis of ‘he who shouts loudest’.

Paragraph 28 rightly states that changes in speed limit should reflect changes in road layout and characteristics. That principle is diluted, however, by the provisions of paragraph 36, which allows for speed limits to be extended under certain conditions beyond the points where they are justified. For speed limits to be respected it is vital that any change in speed limit coincides with, and does not precede, a change in hazard density. The practice of some local authorities to extend speed limits into open country beyond the ends of settlements is counterproductive. Drivers will not slow down if they cannot see a reason for the change in limit, so they may end up travelling faster through the settlement itself than they would do if the limit changed at the point where they enter the built-up area. The second part of paragraph 124 should be strengthened – it is more than “helpful” if drivers can see housing at the same time as the terminal signs, it is essential.

Paragraph 38 allows for the use in some circumstances of transitional speed limits of 40 or 50 mph on the approaches to lower limits in villages or urban areas. Such speed limits are ineffective if the nature of the road suggests to drivers that the national speed limit is appropriate. On single carriageway roads, such ‘buffer’ speed limits are an unnecessary and pointless restriction on drivers leaving the built-up area. Where the terminal signs of a speed limit at the entry point to a settlement have inadequate forward visibility, the ABD would prefer to see countdown boards provided. Although the Department for Transport considers these to be ineffective, countdown boards are provided on the approaches to rural roundabouts and exit slip roads, so the principle must be sound. Perhaps thought should be given to redesigning speed limit countdown boards to make them more conspicuous, or the use of vehicle-activated signs be considered in the limited number of situations where the need arises.

The ABD has serious misgivings about the entire section on rural speed management, especially for single carriageways. Apart from roads that have been recently built or improved, the majority of single-carriageway rural roads have frequently changing alignments, widths and frequency of hazards, such as junctions and sharp bends. The safe speed at which to traverse such roads is constantly changing, therefore, meaning that any speed limit will be either too high or too low most of the time. Local speed limits on such roads may not just be ineffective, they may be counterproductive.

Many people express amazement that narrow, twisty country lanes are usually subject to the national speed limit of 60 mph, when it is obvious that such a speed cannot be achieved safely. In reality, because there are no repeater signs on roads subject to the national speed limit, drivers do not think about the limit at all but simply vary their speed according to the conditions. If a local speed limit is imposed, however, with repeater signs at regular intervals, drivers are constantly being reminded of the limit and, even if only subconsciously, may assume it is safe to drive at that speed. They may thus fail to slow down at hazardous locations, which they might have done if the national speed limit applied.

The increasing use of local speed limits on rural roads typifies a mentality that believes drivers cannot be trusted to think for themselves and must be micromanaged at all times. This is a negative approach in general and simply cannot work with the highly variable nature of the rural road network.

The ABD strongly opposes the proposal to allow consideration of rural 40 mph speed limit zones in places such as national parks. These often cover large areas and traffic levels vary substantially with visitor numbers at different times of the year. It would be quite unreasonable to impose a low speed limit over a large network of roads, along which higher speeds might be safe for much of the time. During peak periods for tourists, speeds will be constrained by traffic volume. Specific problem locations may need remedial treatment on an individual basis, but a blanket approach is unnecessary and unworkable.


The ABD considers the draft circular to be a missed opportunity to reverse the flawed thinking behind current speed limit setting guidance. Despite clear recent evidence from the United States that raising unrealistic speed limits does not generally lead to an increase in accidents, and is much more likely to reduce them, and earlier British experience of the benefits of setting speed limits at the 85th percentile, the mean speed continues to be promoted as the basis of speed limit setting. This will not achieve the draft circular’s stated objectives of achieving self-compliance and bringing speed limits into line with people’s assessment of a safe speed at which to travel.

There does not appear to be a clear understanding in Government of the purpose of speed limits and how they should be used to improve road safety. Until there is a willingness to accept that the policies of the last twenty years are flawed, the misuse of speed limits will continue and road safety will suffer as a result.