Evidence to the Transport Committee's Inquiry into Road Safety
This is the submission of the Association of British Drivers in February 2008 to the House of Commons Transport Committee's inquiry into road safety.
Recommendations for Action
While Britain has traditionally had an enviable road safety record, the rate of progress in reducing casualties has slowed markedly since the mid 1990s. Government claims that its targets for 2010 will be met are based on police figures for seriously injured casualties that are becoming increasingly out of step with hospital records. Fatality figures give a truer indication of progress and these are falling only slowly.
The poor performance in recent years is partly due to lack of investment in the road network, but the ABD believes that much of the decline is a result of focussing road safety policy almost entirely on the claimed contribution of 'speeding' to accidents. This has resulted in widespread reductions in speed limits and a massive increase in speed limit enforcement using cameras. At the same time, police traffic patrols have been cut, so the minority of drivers who commit dangerous offences believe they are less likely to be apprehended. Worst of all, this policy has given the false impression that being a safe driver is just a matter of driving by numbers.
Speed has been made important because it is easy to measure. Dubious research has been used to justify cutting speed limits below the level that decades of international experience has shown to be the optimum for safety - the 85th percentile speed (the speed not exceeded by 85 per cent of vehicles).
The ABD wishes to see mandatory rules applied to highway authorities on setting local speed limits, including reinstatement of the 85th percentile principle. The motorway speed limit should be raised to 80mph, and the 40mph and 50mph national speed limits applying to heavy goods vehicles should be scrapped. Speed limit enforcement should be restricted to stopping drivers at the time and more police traffic patrols should focus on setting an example to other road users, giving advice to those who make minor errors, and deterring the reckless minority.
Traffic calming measures can damage vehicles, injure their occupants, obstruct the emergency services, and may introduce hazards that did not previously exist, leading to more accidents rather than less. Alternative ways of reducing speeds, by changing the appearance of a road to drivers, should be applied more widely.
Allowing local authorities to keep the income from enforcement of waiting restrictions and other traffic regulations has led them to target minor offences rather than attempting to deter those who cause real obstruction or danger. Local authorities should instead be required to bid for funding to carry out enforcement to the level needed for traffic flow and safety, with income reverting to the Treasury.
There needs to be a much greater emphasis on educating and training road users, which has the potential to produce better results than engineering and enforcement combined. Using the roads is an exercise in continuous risk management and it is largely a person's beliefs and attitudes that determine his or her safety. The correct attitudes need to be taught at the earliest possible age, but they must not be about just obeying simple rules. Education should be about teaching the principles within which people are encouraged to think for themselves. Advanced driver training should be encouraged throughout a driver's career.
1.1 The Association of British Drivers (ABD) was formed in 1992 to campaign for a better deal for Britain's motorists. In particular, its founder members were very concerned about the increasing use of technology to enforce driving laws, which threatened to undermine the traditional 'Three Es' approach to road safety (Education, Engineering and Enforcement) that gave Britain the safest roads in the world. Those fears have been more than realised.
1.2 The ABD is a voluntary organisation funded by subscriptions and donations from its members and supporters. It receives no funds from public bodies or private-sector businesses, so is truly independent. The ABD is a member of the Parliamentary Advisory Council for Transport Safety and the National Council of Voluntary Organisations.
1.3 Many of the ABD's active members are from professional or managerial backgrounds, and around 40 per cent of national committee members hold advanced driving qualifications. Malcolm Heymer, who is submitting this evidence on behalf of the ABD, holds a master's degree in Transportation Engineering and has over thirty years' local government experience in the fields of transportation modelling, highway engineering, transport planning and traffic engineering, including road safety engineering.
1.4 Mr Heymer is willing to give oral evidence to the Transport Committee if requested.
2.1 Britain still has one of the best road safety records in the world, but its rate of improvement has slowed dramatically since the mid 1990s. Several other countries now have lower fatality rates, and claims that the government's targets for casualty reduction by 2010 are likely to be met are based on official figures for serious injuries that are highly suspect (Annex B).
2.2 The stagnation in Britain's road safety performance coincided with major cuts in the programme to improve the strategic road network, which is now one of the most underdeveloped in the EU (Annex A). Motorways and dual carriageways have much lower accident rates than single carriageway roads, many of which are now operating beyond capacity, leading to both delays and accidents. Major improvements to the strategic network are essential.
2.3 At the same time as the roads programme was cut, official road safety policy changed to one focussed almost entirely on the simplistic view that speed is the main cause of accidents. This resulted in the deployment of speed cameras to automate speed limit enforcement, and many speed limits have been reduced unnecessarily, leading to large numbers of drivers being prosecuted for what are, in most cases, technical offences creating little or no danger. Worse still, the culture that gave Britain the best road safety record in the world has been undermined. Drivers are now led to believe that they do not have to think for themselves, just stick blindly to a few simple rules. This has been a disaster.
2.4 The majority of accidents are caused by human error, and it invariably requires a combination of factors to come together for an accident to happen. Rather than determine what all these factors are, police investigations are usually aimed at establishing whether a driver has broken the law and can be prosecuted. This approach is hampering the advancement of knowledge about the real causes of road accidents, ensuring that more will occur. An accident investigation body, independent of the police, would encourage drivers to co-operate in establishing the full facts (Annex C).
2.5 While speed limits have a role to play in road safety, they have been given a level of importance they do not deserve because speed is so easy to measure. Even when correctly set, fixed speed limits give no more than a guide to the speed that may be safe on a particular stretch of road, and make no allowance for changing conditions. It is absurd to maintain, therefore, that driving within the speed limit is always safe, or that exceeding it is always dangerous.
2.6 International research over many years has shown that drivers who have the lowest accident involvement travel at speeds that are above the average. That is why the 85th percentile speed (the speed exceeded by only 15 per cent of drivers) is established as the best basis on which to set speed limits. Drivers who travel faster - or slower - than this speed are more likely to be involved in an accident. Those who select a speed around the 85th percentile are exhibiting the greatest skill in matching their speed to the level of hazard density, by observing and assessing the risks presented by all aspects of the road environment on a real-time basis (Annex D).
2.7 It is a tragedy that the Government abandoned the 85th percentile principle in its 2006 guidance to highway authorities on setting local speed limits. By adopting the mean (average) speed instead, the Government has surrendered to those local authorities and interest groups that had ignored the previous guidance, and is undermining the position of other authorities that have fought to retain a degree of objectivity and realism in speed limit setting. By setting speed limits at the mean speed, the safest and most responsible drivers will become criminals if they continue to use their skills in selecting a safe speed for the conditions. This cannot be just or reasonable, as summarised in the following principle adopted by the U.S. State of Arizona:
"The normally careful and competent actions of a reasonable individual should be considered legal."
2.8 Setting speed limits below the 85th percentile level will lead to increased non-compliance and bring all speed limits into disrepute. The ABD considers it essential, therefore, that the 2006 guidance is withdrawn and replaced with rules on speed limit setting that are mandatory for highway authorities. These should include a return to the 85th percentile as the basis on which speed limits are set, and a ban on the use of local speed limits except where needed to indicate significant changes in hazard density. In particular, they should not be used on a whole-route or area basis in the belief that accidents will be reduced regardless of their causes. The research purporting to show a relationship between average speeds and accidents is flawed, because it has not taken due account of the non-linear relationship between accidents and traffic flow.
2.9 The national speed limit on motorways has been in force for over forty years and is widely disregarded. An increase to 80mph would bring the limit into line with many EU countries and close to the optimum speed, calculated by the Transport Research Laboratory, at which overall costs are minimised. Since the effect on actual speeds of a change in speed limit is only a small proportion of that change, raising the motorway limit to 80mph would not lead to a significant increase in speeds, so there would be little change in accidents, emissions or noise levels. There would, however, be a marked decrease in the number of drivers breaking the limit, which would be seen as more realistic.
2.10 The differential speed limits applying to heavy goods vehicles on rural roads are also outdated and, in particular, the 40mph limit on single carriageways, if observed, creates danger by causing frustration in following drivers. This can lead to dangerous overtaking manoeuvres and accidents. This speed limit, and the 50mph limit on dual carriageways, should be scrapped (Annex H).
2.11 Since the introduction of speed cameras, enforcement of speed limits now takes place mainly after the event, with drivers often unaware that they have been caught until a notice of intended prosecution is received two weeks later. This gives drivers little chance to defend the allegations against them, and the automated processing of large numbers of speeding tickets frequently bypasses the legal safeguards intended to protect drivers against wrongful prosecution.
2.12 The ABD believes, therefore, that speed limit enforcement should be restricted to situations where drivers are stopped at the time of alleged offences. This would not only be fairer but would have an immediate effect on a driver's future behaviour. It would mean removing all unattended speed cameras, but these could be replaced with effective measures, such as vehicle-activated signs, to slow drivers on the approach to hazards (Annex H).
2.13 The effectiveness of speed cameras in reducing accidents has been greatly misrepresented. Accidents are random events, and large variations can take place from year to year, especially when studying individual sites or small areas. Most cameras have been installed after an upward blip in accidents, which would be expected to fall again anyway, with or without a camera. The lack of effectiveness of cameras overall can be seen from their failure to have reduced fatalities nationally or within camera partnership areas (Annex G).
2.14 Greater use should be encouraged of speed limits that vary according to time of day, where there are significant variations in risk. An obvious example is outside schools, where the greatest hazards exist only for short periods on weekdays during term time. Even in other locations, the degree to which a fixed speed limit reflects a safe speed will vary according to time of day, weather and traffic conditions, and so on. Accordingly, the police should be encouraged to use discretion in the enforcement of speed limits (Annex H).
2.15 Police traffic patrols have decreased since the widespread deployment of speed cameras, resulting in less likelihood of drivers being observed performing actions that may be more hazardous than breaking the speed limit. There is also less chance of drivers being detected who are unlicensed, uninsured or driving an unroadworthy vehicle. Traffic policing should be made a core police function, therefore, with traffic patrols providing an example to other road users, giving guidance and advice to those who make minor errors, and acting as a deterrent to the reckless minority. It is vital that traffic police are not encouraged by a target-driven culture to prosecute large numbers of drivers for relatively minor offences.
2.16 Advances in technology have improved the safety of modern vehicles, with systems such as anti-lock braking and traction control. There is a serious risk, however, that developments which take partial control of a vehicle from the driver can make them believe that their responsibility has been offloaded to the technology. For this reason, the ABD opposes the use of intelligent speed adaptation (variable speed limiters) in mandatory form. Research should be undertaken into the adverse effects on driver behaviour of the fixed speed limiters already fitted to heavy goods vehicles, coaches and, more recently, smaller goods vehicles. The use of intelligent transport systems (ITS) should be focused on helping drivers avoid delays, and improve safety by monitoring drivers' attentiveness to the driving task (Annexes I and J).
2.17 The spread of traffic calming schemes has done little to reduce casualties overall and has many negative impacts. Road humps cause premature wear to vehicles, and cause discomfort and even injury to vehicle occupants. They delay emergency services, increase fuel consumption, and create more noise and vibration suffered by residents. Other forms of traffic calming often bring vehicles into conflict with one another, raising the risk of accidents, and the features themselves and the signs associated with them are visually intrusive, especially in villages and rural areas. Casualty reduction claims for traffic calming schemes ignore the random nature of accidents and the displacement of traffic the schemes can cause, so they are not reflected in national casualty figures or those of the casualty reduction partnerships (Annex K).
2.18 There should be a freeze on the introduction of conventional traffic calming schemes and the funding arrangements that encourage highway authorities to construct them. Where there is a real need to slow traffic below the speed at which drivers would naturally choose to travel, this should be achieved by changing the visual environment as perceived by drivers. The use of 'psycho-logical' traffic calming and the shared-space concept should be encouraged in appropriate locations, but it is vital that solutions are tailor-made to specific circumstances and they are not applied regardless (Annex L).
2.19 The decriminalizing of waiting restrictions and some other traffic regulations, with local authority enforcement operations financed by the penalties for non-compliance, has led to a huge increase in the issue of penalty charges, often for technical infringements having little bearing on safety or traffic flow. To ensure that the more serious offences are targeted, the financial incentives to issue more penalties must be removed. Local authorities should make an annual bid for funding to carry out enforcement to a level necessary to maintain traffic flow and safety, and to deter the avoidance of legitimate charges. The income from any penalties applied should revert to the Treasury, and there must be no targets for numbers of penalties to be issued (Annex M).
2.20 The role of education in casualty reduction has never received as much attention or funding as enforcement and engineering, yet it has the potential to contribute more than either of those. Most road accidents are caused by errors on the part of one or more road users. While human fallibility is inevitable, the consequences of errors could be mitigated if all road users paid full attention to what they were doing and made allowance for the likely mistakes of others.
2.21 The current driving test concentrates largely on the ability to control a car or motorcycle, knowledge of the law, and hazard awareness. What it does not assess are the attitudes and beliefs of candidates towards driving and safety. Safe driving is essentially an exercise in continuous risk management, so the way a driver thinks is more important than how the car's controls are handled. Instilling the right attitudes concerning road safety needs to start as early as possible, preferably in schools. Everyone needs to be made aware that learning to use the roads safely is a vital life skill (Annex N).
2.22 Road safety education must not mean brainwashing people into slavishly following simplistic rules, since the road environment is too complex for that. People have to think for themselves within a framework of basic principles. Education should focus on teaching those principles and the correct approach to individual risk management on the roads. The driving public should be treated like responsible adults, which most of them are.
2.23 Advanced driver training and testing significantly reduces accident involve-ment among those who undertake it. People who voluntarily take advanced driver training are showing that they have a responsible attitude to using the roads through their desire to improve their skills. More people should be encouraged to take advanced driver training, possibly by introducing a system of positive points for those who successfully complete approved training schemes after the standard driving test. These positive points would help offset any penalties incurred for minor offences.
2.24 Recent increases in drink-related fatalities reflect the reduction in police traffic patrols since the use of unattended speed cameras became widespread. This has led to a perception that there is less likelihood of being stopped than before. The ABD does not condone drinking and driving, and considers that the advice not to drink at all if one is planning to drive is correct.
2.25 The ABD does not support a reduction in the legal blood-alcohol concentration (BAC) level, however, since the current limit is generally seen as reasonable and has a scientific basis. Lowering it would produce more injustices, particularly in 'morning after' situations, when research shows that, for a given BAC, a driver is less impaired when the alcohol level is falling than when it is rising. An increase in drivers who avoid knowingly drinking and driving, but who are punished in situations where they were not unfit to drive, could undermine respect for the law (Annex O).
2.26 An increase in police traffic patrols would enable the current drink-driving legislation to be better enforced, but this should not mean the introduction of random breath testing, which would unnecessarily inconvenience innocent drivers. The police already have the effective power to stop any driver at any time, and they are quite capable of deciding which drivers are likely to be impaired by alcohol. Traffic patrols should use their enforcement powers to improve road safety and apprehend serious and wilful offenders, rather than seek to prosecute large numbers of drivers for the sake of it.
2.27 In conclusion, current road safety policies are negative and confrontational, alienating drivers rather than persuading them. Attempts to reduce accidents by introducing more laws and restrictions, and enforcing absolute offences without flexibility or common sense, are doomed to failure. Treating drivers as if they are all criminals, and issuing large numbers of fixed penalties that give innocent drivers little chance of achieving justice, is unacceptable in a democratic and free society.
2.28 There needs to be a fundamental change in approach. It must be recognised that most drivers are responsible individuals who do not set out deliberately to have accidents, the causes of which are complex. The use of simplistic sound bites like 'Speed Kills' should be abandoned, and the real causes of accidents investigated properly. Most importantly, the education of road users from an early age needs to impart the right beliefs and attitudes to road safety. The criminal law should be reserved for those who wilfully and recklessly endanger the lives of others.
Recommendations for Action
R.1 Due to the unreliability of police reporting of non-fatal injuries, the use of combined killed and seriously injured (KSI) figures should be discontinued in setting future casualty reduction targets, which should focus on the more reliable fatality figure. Casualty reduction partnerships should be required to show fatalities separately in all reporting of road casualty figures.
R.2 The inter-urban motorway and trunk road network requires significant and urgent extra investment to relieve overloaded single-carriageway routes and remove through traffic from towns and villages. This would make a significant contribution to accident savings, as well as reducing congestion, improving journey time reliability, and benefiting local environments and the economy.
R.3 The Government and agencies responsible for improving road safety must recognise that the causes of accidents are complex and cannot be solved with simplistic solutions. There needs to be a fundamental change in thinking, away from the idea that punishing people for human errors and minor technical offences can improve driver performance. The criminal law should be reserved for dealing with wilfully dangerous acts, and the focus of accident investigation should be on establishing all contributory factors, so that further accidents of the same type might be prevented.
R.4 The possibility should be investigated of setting up an accident investigation body, independent of the police, with which drivers could co-operate on a confidential basis. This body would be required to issue annual reports, analysing the results of its investigations to give a greater understanding of accident causation factors.
R.5 Road safety policies must make it clear to all road users that they have a responsibility for their own and others' safety. Road users must be encouraged to think for themselves and not assume that rigid adherence to legal restrictions will ensure their safety.
R.6 There must be a move away from speed limit enforcement as the focus of road safety policy. While speed limits have a legitimate role to play, they can never have more than a minor impact on casualty reduction and their enforcement cannot be a substitute for road safety education and engineering solutions.
R.7 The 2006 guidance on setting local speed limits should be withdrawn and replaced with mandatory new rules, reinstating the 85th percentile principle for speed limit setting. Local authorities should be required to review their existing speed limits and bring them into line within five years. Traffic managers should be required to report progress to the Secretary of State.
R.8 Speed surveys undertaken to establish the 85th percentile speed for speed limit setting purposes must use only the speeds recorded of vehicles unimpeded by those ahead of them, i.e. the surveys must disregard the speeds of drivers who do not have the freedom to choose their speed.
R.9 Local speed limits should only be used to indicate significant changes to hazard density and not on a whole-route or area basis.
R.10 The national speed limit for motorways should be increased to 80mph.
R.11 The 40mph and 50mph speed limits applying to heavy goods vehicles on rural single and dual carriageway roads should be scrapped.
R.12 Speed limit enforcement should be restricted to situations where the driver is stopped at the time of an alleged offence. Fixed speed cameras on the approaches to hazards such as junctions and bends should be replaced with vehicle-activated signs, or other measures to slow traffic.
R.13 The regulations for introducing variable speed limits should be relaxed, so that they can be used more widely in locations where a lower limit is required only at restricted times, such as outside schools.
R.14 The police should be encouraged and legally entitled to apply variable thresholds when enforcing speed limits, to reflect changing degrees of risk.
R.15 Traffic policing should be made a core police function, to set an example to other road users, to give guidance and advice to those who make minor errors, and to act as a deterrent to the unlicensed, uninsured or reckless minority. Traffic police should not be encouraged by a target-driven culture to prosecute large numbers of drivers for relatively minor offences.
R.16 Research should be undertaken on the adverse effects on driver behaviour of the speed limiters currently fitted to some goods vehicle and coaches.
R.17 Any form of technology that takes responsibility away from the driver, such as intelligent speed adaptation in mandatory form, should be prohibited. The use of intelligent transport systems should be focused on helping drivers avoid delays, and improving safety by monitoring drivers' attentiveness.
R.18 There should be a freeze on the introduction of road humps and obstructive traffic calming schemes, and the funding arrangements that encourage them.
R.19 New guidance should be provided to local authorities on alternative methods of reducing traffic speeds where necessary and improving the environment for all road users. Schemes based on the shared-space concept should be encouraged where appropriate.
R.20 The income from penalties for breaches of waiting and other restrictions should no longer fund enforcement operations. Authorities should instead make an annual submission to finance the level of enforcement needed for traffic flow and safety reasons, or to deter the avoidance of parking or other reasonable charges.
R.21 Local authorities should be required to review their traffic restrictions, to ensure that they are still needed for legitimate traffic flow and safety reasons.
R.22 Road safety education should start as early as possible and should teach the correct approach to individual risk management, based on people thinking for themselves rather than always expecting to be told what to do.
R.23 There needs to be recognition that the right attitudes and beliefs in road users are essential for safe use of the roads, and that unnecessary distractions should be avoided when driving, cycling or walking.
R.24 To encourage people to take advanced driver training throughout their careers, a system of positive points should be considered.
R.25 There should be no reduction in the legally permitted blood-alcohol level for driving, since this would increase the number of drivers being penalised for morning-after offences and undermine respect for the law.
R.26 Police should not be given specific powers to carry out random breath tests to detect drink-drivers, as they already have adequate powers to stop and test drivers whom they suspect of having consumed alcohol.
R.27 Reinstatement of police traffic patrols, plus inclusion of advice on the dangers of drink (and drug) driving in road safety education programmes, are the best ways of tackling these problems.
Annex A: Background
Annex B: Casualty reduction targets and the reliability of casualty figures
Annex C: Why are road safety policies failing?
Annex D: The value of speed limits
Annex E: The effect of speed on accidents
Annex F: The effect of speed limits on speed and accidents
Annex G: The enforcement of speed limits
Annex H: A new policy on speed limits and their enforcement
Annex I: Speed limiters
Annex J: Intelligent transport systems (ITS)
Annex K: Traffic calming
Annex L: Psychological traffic calming
Annex M: Enforcement of waiting restrictions and other traffic regulations
Annex N: Education and training
Annex O: Reducing deaths and injuries from drink-driving
Annex A: Background
A.1 Britain has long had an enviable road safety record internationally. That is still the case, but in recent years the UK's record, measured by road deaths per 100,000 population, has improved much less than that of several other developed countries.1 The number of people killed in road accidents per year more than halved between 1965 and 1994, but only fell by a further 7 per cent over the following eleven years.
A.2 The reduction in road deaths during the 30-year period from the mid 1960s can be attributed to several factors, but the dominant ones are better vehicles and better roads. Improvements in vehicle engineering have made it easier for drivers to avoid accidents in the first place (active safety) and reduced the risk of injury to vehicle occupants and other road users when an accident does occur (passive safety). Motorways and dual carriageways, with their limited access and elimination of conflicting turning movements, have a much lower accident rate than all-purpose roads. Development of the trunk road network, therefore, has helped reduce accidents as well as benefiting the economy. Many high-risk accident sites have been eliminated by small-scale road engineering schemes.
A.3 So why has the rate of improvement slowed since the mid 1990s? Vehicle engineering has continued to make great strides in both active and passive safety. New cars today are almost universally fitted with anti-lock braking systems, and traction and stability controls are becoming more common. Improvements in passive safety include more effective crumple zones and the widespread fitment of air bags. Vehicle designs are also taking more account of the need to minimise injuries to pedestrians in the event of a collision.
A.4 It is a different story with the road network. In 1965 there were 566 km of motorway in Britain and this figure had risen to 3,242 km by 1994. Since then, less than 300 km have been added.2 In relation to the size of its economy, the UK's motorway network is woefully inadequate: with fewer than 2 kilometres of motorway for every billion dollars of GDP in 2003, it is less than half the EU average.3 This failure to invest adequately in the national road network is undoubtedly a factor in the reduced rate of fall in road deaths, as traffic levels continue to rise on roads with accident rates higher than those of motorways and dual carriageways. Much of the funding that would previously have been spent on improvements to high-risk accident locations is now allocated to schemes that obstruct drivers or take road space from them.
A.5 Reduced investment in the road network is unlikely, however, to be the only factor - or even the main one - affecting road safety in the last decade. To understand why, it is necessary to consider why Britain has historically had a better road safety record than most other countries. It cannot be due to differences in the vehicle fleet, since most cars in use around the world are made by the same multi-national companies. Nor is it due to the UK having a better road system since, as already demonstrated, most other developed countries have more extensive motorway networks.
A.6 The main reason for international differences in accident rates is probably cultural, demonstrated by different attitudes between countries towards road safety. These differences are impossible to quantify, but Britons have traditionally shown respect for others and for the rule of law - provided that laws, and their enforcement, are seen as just and reasonable. Since the mid 1990s, however, enforcement of road traffic laws has focused increasingly on those offences, such as speeding, that are easy to detect. At the same time, many new restrictions have been introduced, including lower speed limits.
A.7 These developments, coupled with use of the latest enforcement technology, have led to an explosion in the numbers of drivers being prosecuted. This has resulted in increasing resentment by drivers penalised for what are often seen as 'technical' offences. The narrow focus on speed and enforcement of speed limits over the last decade has done a great deal to undermine respect for traffic laws and to damage Britain's road safety culture. In the ABD's opinion, a major change of direction in road safety policy is required.
Annex B: Casualty reduction targets and the reliability of casualty figures
B.1 Since casualty figures are used to measure the effectiveness of road safety policies, it is vital that they are reliable. The number of people killed in road accidents each year has traditionally been seen as the indicator of the national trend, as it is generally accepted as the most accurate. It is also the only one that can be used to give a realistic international comparison, since the level of reporting of less severe accidents varies widely between countries.
B.2 In 2000 the government set new targets for road casualty reductions to be achieved by 2010.4 Instead of setting a separate target for fatalities, however, the targets now are to achieve reductions in the combined totals of killed and seriously injured (KSI) casualties.
B.3 The use of fatality figures as a measure of road safety performance can be problematic when looking at small geographical areas or individual road improvement schemes, since fatalities are, thankfully, rare. This means they can be subject to large random variations from year to year. With roughly ten serious injuries for every fatality, the use of combined KSI figures might be thought to provide a more reliable measure of road safety trends.
B.4 In practice, by combining serious injuries with fatalities, the figures are susceptible to under-reporting of non-fatal injury accidents. It has always been recognised that police figures for serious and slight casualties do not reflect the total numbers. This would not be a problem in monitoring trends if the scale of the under-reporting remained constant over time. There is mounting evidence, however, that this is not the case and official figures are giving a distorted view of progress towards achieving the targets.
B.5 An analysis of historical accident data5 shows how the trend has changed. In 1950 there were just under 10 serious injuries for each fatality. By the mid 1960s, when accident numbers peaked, the ratio of serious to fatal injuries was over 12:1 and by 1985 it was nearing 14:1. These changes show fatalities falling as a proportion of the KSI total and are consistent with the improving passive safety of vehicles and advances in medical treatment.
B.6 Since the late 1980s, however, the trend has reversed and by 2004 there were fewer than 10 serious injuries recorded for each fatality - a level similar to 1950. This is illogical, given the continued advances in passive safety and medical care. It suggests that either under-reporting of serious injuries is getting worse or injuries are being misclassified as slight instead of serious.
B.7 Two reports published in 2006 6 7 compared police figures for serious casualties with those derived from records of hospital admissions. These show that while the official figures, based on police reports, had fallen between 1996 and 2004 at an average rate of 4.4 per cent per year, those based on hospital records had fallen by just 0.2 per cent per year. If the hospital figures are correct, this means that the reduction in serious injuries over that eight-year period has been negligible, so claims about being on course to meet road safety targets by 2010 are unfounded.
Annex C: Why are road safety policies failing?
C.1 With the exception of a tiny minority of people who commit deliberately reckless acts, the majority of road accidents are caused by human error; hardly anyone sets out with the intention of harming themselves or others. Despite this, road safety policy is largely focused on bringing legal sanctions to bear on those who break an ever-increasing number of rules and regulations. When an accident happens, police investigations are more concerned with the possibility of bringing criminal charges than to establish all the contributory factors and how similar accidents might be avoided in the future.
C.2 This is not a new phenomenon; it was the central theme of the seminal book on road safety Road Accidents: prevent or punish? by the late John Leeming, a former county surveyor and pioneer road safety researcher, published in 1969.8 Leeming argued that drivers involved in accidents are unlikely to be completely open with police investigators for fear of incriminating themselves. Factors that may have contributed to the accident, such as a confusing road layout, may go undiscovered and further accidents will occur. The ABD believes that a body should be set up to investigate the causes of road accidents, independent of the police, with which drivers could co-operate in confidence, much the same as the 'no blame' reporting system used by the world's airlines for precisely the same reasons. By publishing regular analyses of these investigations, a greater understanding of the contributory factors to accidents would be obtained.
C.3 Hardly any accidents have a single cause; it almost always requires a combination of factors to come together for an accident to happen. Even when the 'cause' of an accident appears to be obvious, there is often a more important underlying factor at work. For example, if a driver enters a bend too fast and leaves the road, the accident will be attributed to excessive speed, but the real cause was the driver's failure to read the road properly and make a correct estimate of a safe speed for negotiating the bend. The driver's error may, in turn, have been due to inexperience, fatigue, or the road alignment being deceptive, such as the bend tightening part way round.
C.4 The causes of accidents, therefore, are not simple and nor can the solutions be. Yet the recent trend in road safety policy has been based on a simplistic view that punishing more drivers for infringing a greater number of motoring laws will somehow reduce casualty levels. It is clearly not working and, indeed, the worsening trends suggest that current policy is positively harmful.
C.5 An example of the muddled thinking behind this legislative onslaught can be found in the Road Safety Act 2006, which creates a new offence of causing death by careless driving.9 It is claimed that this new offence is needed because of the difficulty of securing convictions for causing death by dangerous driving; the criteria that define careless driving are less onerous to prove. This line of reasoning exposes the desire to seek retribution rather than to prevent accidents happening in the first place.
C.6 Even the best and most conscientious driver makes a mistake occasionally, but whether an accident results - and the severity of any injuries - depends on the chance of other factors coming together at the same time. So any driver could face a prison sentence if a simple human error leads to someone's death. It is absurd to believe that the threat from this new law and others like it can turn human beings into perfect drivers who will never make mistakes.
C.7 It is impossible to commit a burglary or a robbery inadvertently, but most road accidents are caused by errors rather than deliberate intent to do harm. While the worst forms of dangerous behaviour on the roads undoubtedly need to be dealt with through the courts, there needs to be a fundamental rethink of the extent to which the criminal justice system can be expected to improve the standard of the majority of drivers. As John Leeming observed, the greatest deterrent to having an accident is the accident itself. Of course, if a driver's mistake results in an accident, then those affected are rightly entitled to seek compensation through civil action.
C.8 Nowhere is the desire to punish more obvious than in the enforcement of speed limits. Exceeding a statutory speed limit is a criminal offence, and modern technology enables speed limit offences to be detected and processed in large numbers with minimal police involvement. Yet the law on speeding requires a driver to walk a fine line between legal and criminal behaviour that is separated by as little as one mile per hour. This requires drivers to give a significant amount of their attention to monitoring their speed closely, distracting them from observing other things happening around them that may be more important to safety.
C.9 Speed limit enforcement has become a major growth industry in the last decade, not just by providing fine revenue for the Treasury but in supporting the growth of camera partnerships and the manufacturers of speed detection equipment. The focus on speed limits is justified by widely publicised claims about the contribution of speeding to accidents, the reductions in accidents that lower speeds could bring, and the reductions that have been achieved by the use of speed cameras. The validity of those claims will be examined, after first showing how speed limits should be used in the pursuit of road safety and their effect on driver behaviour.
Annex D: The value of speed limits
D.1 Apart from a brief period in the early 1930s, speed limits have been imposed on public highways in Britain for as long as there have been mechanically propelled vehicles. Their use is so well established that few people ever stop to consider their value as an aid to road safety. Yet fixed speed limits have their limitations, as drivers are faced with constantly changing conditions: changes in carriageway width, and in horizontal and vertical alignment; the presence of junctions, accesses and on-street parking; pedestrian and cyclist activity; all of these and more must be taken into account in selecting a safe speed at which to travel. On the same section of road, different speeds may be appropriate at different times of the day or night, on different days of the week and at different times of the year, and as weather conditions change.
D.2 Even where speed limits are set reasonably, therefore, they cannot provide more than an approximate guide to what might be a safe speed under average conditions: at some times or locations the speed limit might be too high; at others it might be too low. Furthermore, speed limits in Britain are by statute set at intervals of 10mph and, to avoid driver confusion, frequent changes in speed limit are discouraged. Given all these factors, it is clear that speed limits are very blunt tools and that drivers should not assume that it is always safe to drive up to them - equally, it is fatuous to assert that it is always dangerous to exceed them.
D.3 Despite the limitations of speed limits, when set correctly they provide a positive contribution to road safety by helping inexperienced drivers to judge what is safe, by warning drivers of expected hazard density, and by providing a reference level for prosecution of those who drive recklessly. Recently, however, speed limits have become over-emphasised in terms of their benefits.
D.4 Where speed limits are set they will always represent a compromise, so it is important that they are applied to lengths of road that are reasonably consistent in character. These conditions are most often met in urban areas, where speeds themselves and the spread of the speed distribution are generally low. In rural areas it is rare for roads to have consistent characteristics, except in the case of modern motorways and dual carriageways. Most single-carriageway rural roads have frequently changing alignments, often with individual hazards (difficult junctions, sharp bends, crests, etc) joined by safer sections.
D.5 In these circumstances it is counter-productive to apply a blanket speed limit along the whole length of a road, on the assumption that it will reduce accidents at the hazardous locations: if the speed limit is set at a level dictated by the individual hazards, it will be too low on the safe sections. As explained below, this means it will be ignored by most drivers and the road will become more dangerous, not less. Similarly, if a limit is set that was reasonable on the safe sections, it could lead to the less experienced drivers increasing their speed through the hazardous sections; again, this would create more danger.
D.6 Accident locations on rural roads need to be treated with engineering measures tailor-made for those sites. These could include junction redesign, carriageway realignment, or additional warning signs and markings; inter-active signs that light up when a driver approaches a hazard above a pre-set speed have been found by the Transport Research Laboratory to be very effective and are much cheaper to install and maintain than speed cameras.10
D.7 On any given stretch of road, drivers will travel at different speeds, and the spread of those speeds will take the form of a 'normal' distribution, as shown by the bell-shaped curve in the graph below.11 The vertical line through the crest of the curve represents the 50th percentile speed (the speed below which half of drivers travel). Another vertical line, to the right of the graph, shows the 85th percentile - the speed not exceeded by 85 per cent of drivers.
D.8 The crash-risk curve shows the relative accident risk of drivers in relation to their speed within the speed distribution. This curve is based on the results of studies in various countries of speed and accident involvement.12 It can be seen that, contrary to what might be intuitively expected, accident risk does not rise in a simple relationship with speed but is actually lowest for drivers travelling in the 80th to 90th percentile speed range. At speeds above the 90th percentile, accident risk rises sharply, but it also rises at lower speeds and, indeed, the slowest drivers are at similar risk to the fastest.
D.9 It is because of these findings that the 85th percentile speed has long been recognised by traffic engineers as the optimum level at which to set speed limits. When speed limits are set in accordance with the 85th percentile, it means that the majority of drivers, including the safest, are travelling within the law.
D.10 The rationale for setting speed limits according to the measured 85th percentile is summed up in the following principle adopted by the Arizona Department of Transportation:
“The normally careful and competent actions of a reasonable individual should be considered legal.“13
D.11 Speed limits based on the 85th percentile reduce the spread of the speed distribution, especially by reducing the number of drivers travelling at the highest speeds: since the speed limit has been set in accordance with the actions of the responsible majority, the remaining drivers are more likely to accept it as reasonable. The converse is also true: if speed limits are set below the 85th percentile speed, they are largely ignored, even by the safest drivers; the reckless minority then see that the limit has no respect so they ignore it altogether. Thus speeds can actually increase when unreasonably low speed limits are introduced, and can decrease when speed limits are raised.14
D.12 There are other adverse effects of unrealistically low speed limits. Those drivers who are fastidious about always obeying the law, no matter what, will create a queue of impatient drivers behind them. Eventually, frustration may lead to a driver attempting an unsafe overtaking manoeuvre, putting all road users at risk, including the law-abiding driver at the head of the queue.
D.13 The need for a driver to exercise judgement in adjusting speed to the prevailing conditions is one of the fundamental requirements of driving. Speed limits should be used to assist drivers in that task, not to take responsibility away from them. Overuse of speed limits carries the risk of drivers becoming reliant on external information for adjusting their speed, when the factors governing the safe speed under a particular set of conditions are too complex for any system of speed limits to replace a driver's judgement. This tendency towards 'driving by numbers' is eroding drivers' abilities.
Annex E: The effect of speed on accidents
E.1 While the basic laws of physics dictate that, other things being equal, the higher the impact speed the more serious the consequences of an accident will be, this does not mean that higher speeds automatically lead to more frequent or more severe accidents. Drivers travel faster on roads with fewer of the hazards that provide opportunities for accidents to happen, such as junctions, bends and the dangers associated with built-up areas. At the same time, because there are fewer hazards, roads with higher average speeds have lower accident rates - motorways are our safest roads.15
E.2 Furthermore, except in rare cases - such as a driver falling asleep and crashing without taking any avoiding action - impact speed is not the same as approach speed. Nor is there any fixed relationship between the two speeds, since much depends on the alertness and responsiveness of the driver(s) concerned. For this reason, speeding campaigns are misleading when they quote the percentage of pedestrians killed if hit at various speeds: official casualty figures show that just 2 per cent of pedestrians died who were involved in reported injury accidents in 200516, suggesting that actual impacts take place at very much below approach speeds in most cases.
E.3 Despite the fact that roads with the highest speeds have the lowest accident rates, recent government policy on the setting and enforcement of speed limits is based largely on research by the Transport Research Laboratory (TRL) claiming that reducing average speeds can lead to fewer accidents 17 18 19. One of the most frequently quoted claims of that research is that a one mile per hour reduction in average speeds results in 5 per cent fewer accidents.
E.4 This and similar claims are based on comparisons of accident rates between roads claimed to be of similar type but having different average speeds. The fundamental problem with these comparisons concerns the reasons why roads should have different average speeds if they are, indeed, similar in all other respects. Either there are differences that have not been recognised, making the comparison invalid, or the roads have different traffic flows.
E.5 Accident numbers increase with traffic flow, but not in direct proportion. Thus the accident rate (expressed as accidents per 100 billion vehicle-km) decreases with increasing flow. But speeds also decrease as traffic levels rise, so it is wrong to claim that decreasing accident rates are the result of lower speeds. Rather, both accident rates and speeds decrease due to higher traffic flows.
E.6 Evidence that the claimed relationships between average speeds and accident rates are flawed comes from the TRL itself. In a technical guide for road safety engineers20, it is stated that, because the relationship between accidents and vehicle flows is not a linear one, roads with very different flow levels should not be studied together.
Annex F: The effect of speed limits on speed and accidents
F.1 Belief in the discredited relationship between average speed and accident levels has led to many traffic authorities reducing speed limits, on the assumption that accident numbers will fall, regardless of their causes. The Department for Transport published new guidance on the setting of local speed limits in 200621, abandoning the 85th percentile as the basis of speed limit setting and replacing it with the mean, or average speed. This will result in many more speed limit reductions, with speed limits being set at a level below that at which the safest drivers would otherwise choose to travel. As a result, respect for speed limits will be undermined further.
F.2 There is a widespread misconception that reducing - or raising - a speed limit by, say, 10mph will result in an equivalent change to average speeds. Drivers adjust their speed, however, largely in accordance with visual clues to the level of danger ahead of them, and the most successful drivers in selecting safe speeds are those who travel in the 80th to 90th percentile range of the speed distribution. If the speed limit lies within or close to this range there will be a good level of compliance. If not, compliance will be low and there will be a greater spread of speeds, increasing accident risk. As noted by the West Virginia Department of Transportation:
“There is a common belief among laymen and even some elected officials that traffic speeds can be lowered by merely posting signs. This is not true. Artificially low speed limits invite violations by responsible drivers. Enforcement of unrealistically low speed limits sets up a 'speed trap' which is poor public relations and causes a loss of respect for traffic law enforcement activities in general.”22
F.3 Research suggests23 that the change in average speed as a result of a change in speed limit will not be more than 25 per cent of that change, i.e. 2.5mph if the speed limit is raised or lowered by 10mph. Reducing speed limits does not, therefore, lead to equally lower speeds but results in increased non-compliance and lack of respect for speed limits generally.
F.4 The greatest opportunity in recent years to study the effects of speed limits on accidents arose with the repeal of the 55mph federal speed limit in the United States in 1995. From that date, individual states were free to raise their speed limits and many did so, increasing them to 65 or 75mph. The U.S. road safety lobby predicted thousands more road deaths annually, but analyses of accident figures 24 25 show that this did not occur. Indeed, many of the states that raised their limits saw greater reductions in accident rates than those that did not.
F.5 A good example is the State of Michigan, where the speed limit on rural freeways has been raised to 70mph. The State Police now testify at legislative hearings that they want 85th percentile posted limits and are strongly opposed to underposted speed limits that increase speed variance, decrease safety, and foster speed traps. In 2006, the fatality rate in Michigan was 27 per cent below the U S. average.26
F.6 One U.S. State, Montana, went further than any other by abolishing daytime speed limits outside urban areas altogether, insisting only that drivers travel at a speed that was 'reasonable and prudent' for the conditions. This situation continued until 1999 when, under pressure from the U.S. Supreme Court, a 75mph speed limit was introduced on strategic routes and 65mph on other rural roads. During the period of no speed limit, road deaths on rural roads in the state fell to their lowest ever recorded, despite an increase in traffic volumes of more than 12 per cent.27 In the first year after the reintroduction of rural speed limits, fatalities rose by over 40 per cent.
F.7 The improvement in safety during the period of no maximum speed limit was most likely due to greater awareness and vigilance among road users, as responsibility for adopting a safe speed was passed back to drivers. With the reintroduction of speed limits, drivers once again felt absolved from that responsibility.
F.8 One of the claimed justifications for the massive increase in speed limit enforcement in recent years is that 'speeding' contributes to a significant proportion of accidents - a figure of one-third of accidents has been quoted frequently, and sometimes a higher proportion. There was never any hard research backing up these claims, which were also somewhat vague about the difference between exceeding a speed limit and driving at an inappropriate speed for the conditions.
F.9 Research published in 200628 at last made an objective assessment of the role of speed in road accidents, both in excess of a speed limit and within the speed limit but too fast for the conditions. It found that exceeding the speed limit was a factor in just 5 per cent of all accidents and 12 per cent of fatal ones. Driving too fast for the conditions contributed to a further 10 per cent of total and 14 per cent of fatal accidents. It is clear that enforcement of speed limits does nothing to prevent accidents in the latter category.
Annex G: The enforcement of speed limits
G.1 Prior to the authorisation of cameras for enforcement purposes, speed limits were enforced by police officers either at the roadside, using portable speed detection equipment, or by police traffic patrols. Drivers were usually stopped at the time of an alleged offence and shown the evidence against them. This meant that there was rarely any doubt about who was driving at the time, and the driver would have a good idea of whether the claimed speed was accurate.
G.2 In law, a driver cannot be convicted of exceeding a speed limit on the evidence of a single witness.29 When drivers are stopped at the time of an alleged offence, the primary evidence against them is the opinion of the police officer that they were exceeding the speed limit. The secondary evidence is the reading from an approved speed-measuring device.
G.3 The Road Traffic Act 1991 legalised the use of roadside equipment for the detection of traffic offences.30 The first speed cameras were installed in 1992 but, due to the high costs of camera installations, their numbers were restricted until the cost-recovery scheme was introduced nationally in 2001. This enabled camera partnerships (comprising police, traffic authorities and magistrates courts) to recover the costs of enforcement from the income obtained by issuing fixed penalty notices.
G.4 With unattended speed cameras it is impossible to stop a driver at the time of detection. A notice of intended prosecution is sent to the registered keeper of the vehicle, who is asked to provide details of the driver at the time of the alleged offence. The driver is then sent either a conditional offer of a fixed penalty or, if the measured speed is much higher than the speed limit, a court summons. Most mobile camera enforcement is processed in the same way, even when there is an operator present.
G.5 The move away from stopping drivers at the time of alleged offences, coupled with the financial incentives of the cost-recovery scheme, have led to erosion of the safeguards against drivers being wrongly convicted. There must still be two pieces of evidence that a driver was exceeding the speed limit: in the case of unattended Gatso cameras, for example, the primary evidence is produced by a radar speed meter; the secondary evidence comes from measuring the distance travelled between two photographs taken a known time apart. With attended mobile cameras, the primary evidence is still the operator's prior opinion of speed, backed up by the measurement of an approved device.
G.6 In their desire to process as many speeding tickets as possible, it is clear that the camera partnerships routinely send out notices of intended prosecution (NIPs) without checking the secondary evidence that would expose false speed readings. With mobile cameras, NIPs are sent on the basis of the secondary evidence alone, with no record being kept of the primary evidence against a driver - the operator's prior opinion of speed. Some partnerships allow civilians instead of police officers to operate mobile cameras, and some devices are operated at such long range that it would be virtually impossible for even an experienced police officer to assess a vehicle's speed by eye.
G.7 These practices show that camera partnerships frequently breach the spirit if not the letter of the law in enforcing speed limits. In addition, a NIP arriving two weeks after the event does nothing to change a driver's behaviour in the intervening period, and the driver may well be unable to recall the circumstances at the time. A driver may not know, therefore, whether the claimed speed is accurate, or even who was driving. Camera partnerships often refuse to acknowledge that there is a defence in law if a vehicle keeper cannot identify the driver at the time of the alleged offence31, and some are reluctant to allow drivers to see the evidence against them unless they opt to go to court - and risk higher fines and increased penalty points if found guilty.
G.8 A further cause of resentment is the knowledge that drivers who have not registered their ownership of a vehicle with the DVLA cannot easily be traced and can offend with impunity. Such people are also more likely than most to drive vehicles that are uninsured or unroadworthy, or they may not have a driving licence. It is this irresponsible minority that the police should be targeting, but instead it is the generally safe and law-abiding drivers that are penalised by camera enforcement.
G.9 Given these injustices, there is a strong case for restricting speed limit enforcement to situations where the driver is stopped at the time by a police officer and shown the evidence against him or her. This would mean the end of speed camera enforcement in its current form and the mass prosecutions that go with it. By stopping a driver at the time, it is likely that the experience would affect the driver's future behaviour straight away, and the police would need to target their resources at sites where speeding was a proven factor in accidents.
G.10 The justification for allowing a massive increase in speed limit enforcement through the cost-recovery scheme was the claimed effectiveness of speed cameras in reducing speeds and casualties at camera sites.32 The claimed reductions in casualties, however, are largely illusory and are not reflected in overall casualty figures, either within individual partnership areas or nationally. To understand why this is so, it is necessary to examine the process by which speed camera sites have historically been selected.
G.11 The Department for Transport issued rules and guidance for camera partnerships.33 These included criteria for site selection, including a minimum number of killed or seriously injured (KSI) casualties in a 3-year period and a measured 85th percentile speed that is substantially higher than the speed limit.
G.12 The first problem with these criteria is the nature of accidents themselves. These are random events, and the smaller the area studied the greater the chance variation in numbers from year to year. A site selected for camera enforcement on the basis of an upward blip in accidents will almost certainly see a reduction in accident numbers in following years, with or without a camera being installed. This effect, called regression to the mean, accounts for a significant proportion of the casualty reductions attributed to cameras.34
G.13 It has been shown above (para. D.8) that the safest drivers travel in the 80th to 90th percentile speed range and that speed limits should be set as close as possible to the 85th percentile. If the measured 85th percentile speed is significantly higher than the speed limit, it is an indication that the speed limit is unnecessarily low at that point.
G.14 Speed camera 'sites' are not discrete locations but are lengths of road between 0.4km and 20km long.33 Since higher speeds are found on stretches of road with fewer hazards, a camera sited where speeds are highest is unlikely to be where most accidents have occurred. It is not forbidden for partnerships to site cameras in this way and many have done so, encouraged by the former cost-recovery scheme. If cameras were only sited where it is always dangerous to exceed the speed limit, few drivers would be caught and the income generated would not have covered the partnerships' costs.
G.15 Increasingly, the deployment of cameras along substantial lengths of road is justified by the claim that reducing average speeds will reduce accidents, regardless of whether excessive speed was a factor in the accidents that had occurred. As shown above (paras. E.3 to E.6), the research on which that claimed relationship is based was intrinsically flawed.
G.16 As a result of the flawed understanding of the value of speed limits, which has led to them being applied and enforced inappropriately, the ABD believes that a fundamental change of policy is required relating to speed limits.
Annex H: A new policy on speed limits and their enforcement
H.1 One of the first steps should be to withdraw the 2006 guidance on setting local speed limits and reinstate the 85th percentile principle. This should be included in new mandatory rules for speed limit setting, with which traffic authorities would have to comply. Speed surveys to determine the 85th percentile speed should only measure the speeds of unimpeded vehicles. All local speed limits would need to be reviewed and brought into line with the new rules within a specified period of time, say five years.
H.2 The rules should include a ban on setting local speed limits on a general or whole-route basis on rural roads. Local speed limits should only be used to indicate a significant change in hazard density, not as a blanket approach to casualty reduction or to deter use of certain types of road. Individual hazards should be treated with engineering measures, or speeds reduced using methods such as vehicle-activated signs and psychological traffic calming (Annex L).
H.3 In addition to local speed limits, some national speed limits need to be revised. The motorway speed limit of 70mph has been in force for over 40 years and 56 per cent of cars exceed it.35 The Transport Research Laboratory has calculated that the optimum speed for motorways is 78mph, balancing travel time savings against vehicle operating costs and accident costs.36 There should be an immediate increase in the national speed limit on motorways, therefore, to 80mph. The effect on actual speeds would be quite small (see para. F.3), so increases in vehicle emissions or noise levels would be negligible. Raising the speed limit would bring it much closer to the 85th percentile, meaning that a far smaller proportion of drivers would be acting outside the law.
H.4 Other speed limits long overdue for review are those that apply to heavy goods vehicles on rural roads. These vehicles are restricted to 50mph on dual carriageways and 40mph on single carriageways. The 50mph limit is effectively redundant, as it is only 6mph below the governed maximum speed of the vehicles. The police are unlikely to enforce it, therefore, and 86 per cent of articulated vehicles exceed it.35
H.5 The 40mph speed limit on single carriageways is also widely ignored, with 77 per cent of articulated goods vehicles exceeding it. Some camera partnerships are using vehicle detectors to target heavy goods vehicles on roads where the national speed limit for cars is 60mph. Stricter enforcement of the 40mph limit is detrimental to safety, as it leads to long queues of cars stuck behind lorries, leading to frustration and potentially dangerous overtaking: far more car drivers will attempt to overtake a lorry travelling at 40mph than at 56mph. With the improved performance and braking abilities of modern goods vehicles, these differential speed limits are obsolete and should be scrapped.
H.6 Speed limit enforcement should be restricted to situations where the police stop (or attempt to stop) a driver at the time. This would mean the removal of all unattended speed cameras. Where speed cameras are sited to reduce speeds on the approach to hazards, they should be replaced with vehicle-activated signs or other measures to slow drivers without prosecuting them after the event.
H.7 Since fixed speed limits make no allowance for varying levels of risk, it should be made easier to introduce variable speed limits in places where a low speed is appropriate only at restricted times, such as outside schools. For the same reason, the police should be encouraged and legally entitled to apply variable thresholds for deciding when to prosecute those exceeding speed limits. The guiding principle should be the promotion of road safety, not enforcement for its own sake.
H.8 Traffic policing should be made a core police function and the resources provided to enable an increase in patrols by trained traffic officers. Their purpose should be to set an example to other road users, to give guidance and advice to those who make minor errors - and to act as a deterrent to the unlicensed, uninsured or reckless minority. They should not be encouraged by a target-driven culture to prosecute large numbers of drivers for relatively minor offences.
Annex I: Speed limiters
I.1 Speed limiters have been fitted to heavy goods vehicles, buses and coaches for several years, and their use has recently being extended to smaller goods vehicles and minibuses. The devices currently in use are fairly crude and simply limit a vehicle's top speed to the maximum permitted for that class of vehicle anywhere, for example 56mph (90 kph) for goods vehicles over 7.5 tonnes. They have no effect, therefore, on a driver's control of the vehicle on roads with a lower speed limit.
I.2 A new form of speed control device, called Intelligent Speed Adaptation (ISA), is now being promoted as a possible fitment to all vehicles in the future. An in-vehicle 'black box' would contain a database of all the speed limits in the country, while GPS technology would track the vehicle's position and speed. If a driver failed to slow sufficiently when approaching a lower speed limit, the ISA device would react in one of three ways: remind the driver of the speed limit but take no action if the driver ignored it; control the speed of the vehicle to the speed limit by reducing engine power and applying the brakes, but with the driver being able to override the system; or control the vehicle's speed to the speed limit, with no override option.
I.3 The ISA system would affect a vehicle, therefore, at all times and in all speed limits. If used in advisory mode, it could be of some benefit, as drivers sometimes exceed speed limits inadvertently. Once such a system had been fitted, however, there can be little doubt that the override option would be removed. It has been claimed that the universal fitment of a mandatory ISA system could reduce injury accidents by 20 per cent.37 But this theoretical claim is based solely on the assumed relationship between average speed and accidents that has already been shown to be false.
I.4 It is also necessary to take into account the adverse effects that ISA could have on driver behaviour. The overuse of speed limits already has the subconscious effect on some drivers that they feel absolved of responsibility for choosing a safe speed, as the thinking has been done for them. By making it impossible to exceed speed limits, that effect would be much greater. Simulation trials have confirmed this, with those 'driving' an ISA-equipped car being more prone to red-light running, driving too close to the vehicle in front, and driving too fast on a foggy motorway, than when they have full control of the car's speed.37
I.5 It is only necessary to observe the effects on road behaviour of the speed-limited vehicles that already exist, to understand how disastrous the universal fitment of limiters would be. It is common to follow a lorry trying to overtake another on a two-lane dual carriageway or motorway. With a speed differential between the two vehicles of one mile per hour or less, other traffic can be held up behind the two-abreast convoy for several miles - with the following queue of vehicles invariably travelling too close together.
I.6 Even when they are not overtaking, headways between lorries are often dangerously short. Where a vehicle is able to travel only marginally faster than the one ahead of it, the driver will frequently wait until his lorry has almost caught the one in front before pulling out to overtake, to minimise the time spent in the outer lane. These short headways have the potential to lead to multi-vehicle accidents.
I.7 Also, in a vehicle travelling at a constant speed for long periods of time, boredom and fatigue set in and reduce a driver's ability to react to any unforeseen emergency, especially in light traffic conditions or at night. In extreme cases, a driver might fall asleep, leading to the vehicle leaving the carriageway. If cars were also fitted with speed limiters, ensuring that they all travelled at virtually the same speed, the problems would inevitably become much worse.
I.8 The proponents of ISA would doubtless argue that the headway problem could be overcome if vehicles were also fitted with a form of Adaptive Cruise Control (ACC), which senses the distance of a vehicle from the one in front and maintains a safe headway from it. But this would take even more control and responsibility away from drivers, reducing still further their ability to think for themselves. Research suggests that too many driver 'aids' lead to drivers becoming psychologically under-loaded, reducing their ability to deal with an emergency situation.38
I.9 Rather than consider the further extension of speed limiters or the introduction of ISA, priority should be given to researching the adverse impact on road safety that speed limiters are having already. The use of technology should be limited to ways in which drivers can be helped to use the roads as efficiently and safely as possible, without taking control of their vehicles from them. It is these positive aspects of new technology that will now be considered.
Annex J: Intelligent transport systems (ITS)
J.1 This is an umbrella term for technologies that can help drivers complete their journeys quickly and safely. It includes the provision of real-time information at the roadside, such as the variable message signs appearing on some motorways to warn of accidents or other delays, and details of spaces available in town centre car parks. Future developments could include feeding information on congestion to in-vehicle satellite navigation systems, which could then advise a change of route to avoid delays.
J.2 In terms of road safety, ITS has the potential to assist drivers provided it does not take away a driver's control of the vehicle. Most accidents are caused by road user errors and many of these are due to insufficient attention being given to the task in hand. In its research on contributory factors to road accidents, for example, the Department for Transport found that 'failed to look properly' was a factor in 32 per cent.28 Canadian research into driver distraction found evidence that in almost 80 per cent of collisions, the driver looked away from the road ahead just prior to the onset of the crash.39 The development of in-vehicle sensors that could monitor a driver's eye and head movements for signs of drowsiness or inattention, and give audible or visual warnings, could therefore provide a valuable contribution to accident reduction.
Annex K: Traffic calming
K.1 One of the most negative aspects of modern road engineering is the spread of 'traffic calming' schemes, especially road humps. While the latter are effective in reducing vehicle speeds, they have many adverse effects. These include premature wear of vehicle suspension and steering components, and damage to exhaust systems. Drivers (and their passengers) with back problems may suffer discomfort, or worse, from repeatedly travelling over humps. This is also a concern for ambulance crews transporting people with suspected spinal injuries, or attempting emergency treatment on the way to hospital.
K.2 Delays to emergency services' vehicles as a result of road humps are also of serious concern. In 2003 the chairman of the London Ambulance Service estimated that 800 victims of cardiac arrest in London die for every minute of delay caused.40
K.3 Residents of streets where road humps have been installed suffer from the increased noise of vehicles accelerating and braking, often in a lower gear than if they were travelling at a constant speed. They may also suffer from vibration as vehicles, especially lorries, drive over the humps. The extra traffic signs required, as well as the humps themselves, detract from the street scene.
K.4 Forms of traffic calming other than road humps usually attempt to reduce traffic speeds by bringing drivers into conflict with one another. Chicanes and pinch points limit road width, forcing one driver to give way to another from the opposite direction. This sometimes leads, however, to drivers accelerating in an attempt to get through the obstruction first, thus increasing speeds rather than reducing them. Just as worryingly, it causes drivers to focus further ahead, instead of watching out for hazards closer in front of them, as they should be doing in a built-up area.
K.5 Traffic calming features that restrict traffic to a single lane introduce hazards that did not previously exist. When badly sited, such as on or just after a bend, they can be particularly dangerous. Cyclists may find themselves squeezed by motor vehicles if they cannot ride around the obstructions. In order to mitigate these dangers, the forest of traffic signs required is ugly and intrusive, especially in attractive villages.
K.6 Given all the negative impacts of conventional traffic calming, it seems astonishing that so many schemes have been installed in recent years. Road humps first came to widespread public attention as a means of discouraging criminal 'joy riders' from racing stolen cars through deprived housing estates. Residents of other areas then started demanding them as a perceived solution to 'speeding' by drivers generally.
K.7 While there are undoubtedly residential streets and village centres where some drivers travel at inappropriate speed, much of the popular perception of speeding stems from increases in traffic flow rather than speed itself. As vehicles pass a person's home with increasing frequency, the impression can be gained that the traffic is travelling faster, when it is not.
K.8 Although traffic levels have increased over the years, residents of some areas have suffered more than others due to the policies of both central and local government. Reduced investment in the road network since the early 1990s has led to many bypass schemes being shelved, leaving towns and villages to suffer worsening environmental impacts from through traffic. In cities, delays to traffic on main roads as a result of road-space reallocation schemes have led drivers to seek alternative routes through residential areas, fuelling demands from residents for measures to discourage them. Road humps installed in the worst affected streets simply displace the problem, leading to demands for action in more roads - and so the vicious circle continues.
K.9 The exponential increase in numbers of traffic calming schemes has been assisted by government funding policies. Misleading claims about accident reductions (ignoring displacement of traffic and regression to the mean - see para. G.12) coupled with the relatively low cost and prevailing anti-car culture, have actively encouraged councils to install more of these schemes.
K.10 There are signs, however, that the tide of opinion is turning against intrusive traffic calming, as the proliferation of schemes exposes more people to their negative effects. Residents who are initially in favour of road humps often change their minds when they have to live with them. Some highway authorities have stopped installing road humps and others have even begun to remove them. One example is the London Borough of Barnet, which is not just removing humps but is also freeing up its main traffic routes to reduce the temptation for drivers to cut through residential streets. This is despite strong opposition from Transport for London, with threats to withhold future funding and dire predictions of more accidents - which have not materialised.
K.11 The current approach to traffic calming in Britain is symptomatic of the way drivers are treated in general - as incapable of acting responsibly without constant regulation and coercion. In fact, the majority are reasonable people and resent being patronised, bullied and demonised. Several European countries are turning away from intrusive, physical traffic calming and are instead finding ways to reduce speeds through a range of measures that can loosely be termed 'psychological' traffic calming. Some progressive local authorities in the UK have also begun to embrace the concept. Its potential benefits will now be discussed.
Annex L: Psychological traffic calming
L.1 The speed at which drivers travel along a particular road is determined largely by the visual environment ahead of them, which informs their perception of the level of danger. Thus drivers will travel faster along a wide, straight road through open fields than on a narrow, twisty street through a village centre. Most drivers are capable of adjusting their speed to a safe level reasonably well for changing conditions. There are circumstances, however, where the speeds chosen by drivers, while not necessarily unsafe, may seem excessive to residents, pedestrians or cyclists.
L.2 To get drivers to reduce their speed in these situations, without resorting to physical traffic calming, requires the visual environment to be changed to create a perception of greater risk. This can be achieved in a variety of ways, including narrowing the carriageway, tree planting, lowering lamp columns, and additional road signs and markings. The effectiveness of a range of these measures has been investigated by the Transport Research Laboratory.41 42 While most produced some reduction in traffic speeds, they were generally not as effective as 'traditional' traffic calming.
L.3 Some measures, such as extra signs and road markings, add to visual intrusion in settlements and rural areas. An alternative and more radical approach, which is becoming popular in some European countries, is to remove all but the most essential traffic signs, road markings and other street furniture in order to blur the distinction between areas reserved for vehicles and other road users.43 This 'shared space' concept breaks down the traditional idea of one type of road user having right of way over another and makes everyone take more care. As a result, traffic speeds are reduced and streets become more attractive places for everyone.
L.4 Shared-space schemes have yet to gain general acceptance in the UK, but one outstanding example is High Street, Kensington. This main east-west route to central London, which is also a major shopping destination, carries up to 3,000 vehicles per hour in parts and about 120,000 pedestrians during shopping hours on weekdays.44 The scheme that has been implemented has, among other things, removed pedestrian guardrails and all but the most necessary street clutter, straightened kerb lines and provided more cycle parking.
L.5 While the scheme has rejected conventional wisdom that vehicles and pedestrians must be segregated for safety reasons, injury accidents have fallen by 43 per cent, compared with 17 per cent for London as a whole in the same period.44 Even so, other local authorities may be reluctant to pursue similar schemes due to fear of legal action if accidents occur.
L.6 It may be necessary, therefore, to provide new guidance and support to local authorities to enable them to pursue shared-space schemes in appropriate locations with confidence. Meanwhile, there should be a freeze on the installation of conventional traffic calming schemes and the funding arrangements that encourage them.
Annex M: Enforcement of waiting restrictions and other traffic regulations
M.1 The Road Traffic Act 1991 paved the way for local authorities to take over the enforcement of waiting restrictions from the police. Decriminalised parking enforcement (DPE) was initially introduced in London and has since spread to many other areas. The initial reason for giving enforcement powers to local authorities was that the police lacked the resources to enforce restrictions needed for safety reasons and to keep traffic moving.
M.2 The main objection to DPE in its current form is that each authority's scheme is required to be self-financing. This means that the income from penalty charge notices (PCNs) must at least cover the operating costs. As a result, enforcement priorities are governed by the need to issue the maximum number of PCNs rather than to achieve compliance with regulations where they are most needed. This has led parking attendants to target minor technical infringements, rather than using their presence to deter obstructive or dangerous parking.
M.3 The Traffic Management Act 2004 has allowed local authorities to take over enforcement of some other traffic regulations, such as box junctions, banned turns and bus lanes, with increased use of camera evidence. Again, the proceeds from penalties issued are used to fund the enforcement operation, leading to over zealous targeting of minor infringements.
M.4 As well as leading to distorted priorities, allowing local authorities to keep the income from their enforcement activities results in a lack of impartiality when dealing with appeals against incorrect or unfairly issued penalties. Drivers may receive penalty notices as a result of errors by parking attendants or processing staff, through out-of-date or incorrect information on the DVLA database, or as a result of the criminal 'cloning' of vehicle identities. Legitimate appeals in such cases are often met with hostility and drivers forced to prove their innocence - the complete opposite of natural justice.
M.5 To restore justice and fairness to the enforcement of parking restrictions and other traffic regulations, the financial incentives to issue more penalties must be removed. It is perfectly legitimate for local authorities to retain income from charges for services such as public car parking, but enforcement against evasion of such charges and breaches of other regulations must not be financed from the penalties applied. Instead, local authorities should be required to make an annual bid for funding to carry out enforcement to a level necessary to maintain traffic flow and safety, and to deter the avoidance of parking or other charges. The income from any penalties applied should revert to the Treasury, and there must be no targets set for the numbers of penalties to be issued.
M.6 Local authorities should also be required to review their waiting restrictions and the other traffic regulations that they enforce, to ensure that they are still needed for legitimate traffic flow and safety reasons. Any that are no longer justified should be removed.
Annex N: Education and training
N.1 Of the 'Three Es' of road safety, education has long trailed behind engineering and enforcement in emphasis and funding, and is often treated as an afterthought. Yet it has the potential to deliver far greater improvements in casualty reduction than either of the other two.
N.2 The basic driving test in the UK assesses a driver's mechanical handling of their vehicle, knowledge of the Highway Code, ability to spot potential hazards, and completing a short test route to the satisfaction of the examiner. All these are important, but the test does not assess the most important thing of all - attitude. Safe driving is essentially an exercise in continuous risk management, so the way a driver thinks is more important than how the car's controls are handled.
N.3 The overwhelming majority of road accidents are caused by human error on the part of one or more road users. Some of those errors are caused by lack of knowledge or experience, but many are due to distraction or inattention. To many people, using the roads is just a matter of getting from A to B and walking, cycling and driving are seen as secondary activities. As a result, people do not give them sufficient attention.
N.4 Road safety policies that focus on the enforcement of absolute laws create the impression that sticking to a few rules is more important than taking personal responsibility for minimising risk. There is little incentive to develop the attitude and approach to driving that enables the best drivers to complete their journeys in safety. Worse still, the implication that little effort is required to be safe on the roads reinforces the view that there is no need to give full attention to the task. Thus people can be led to believe that it is safe to use a mobile phone, shave, apply make-up, read, or eat and drink at the wheel - provided they do not break the speed limit.
N.5 There is no doubt that advanced driver training and testing significantly reduces accident involvement among those who undertake it. The Transport Research Laboratory found that drivers who passed the Institute of Advance Motorists' test had 25 per cent fewer accidents than those who failed it45. One fleet manager recorded an 85 per cent reduction in accident rate among company drivers after implementing a driver training and risk management regime, and benefited from a 30 per cent cut in insurance costs46. When advanced drivers do have accidents, they are less serious in terms of injuries and damage, are less likely to involve pedestrians or cyclists, and the advanced driver is less often at fault47.
N.6 People who voluntarily take advanced driver training or testing are showing that they have a responsible attitude to using the roads through their desire to improve their skills. Yet there is little incentive for the majority of drivers to take further training, apart from the possibility of a reduction in insurance premiums. Most people regard passing the standard test as the end of their driving education, when really it is just the end of the beginning.
N.7 One way of encouraging more people to take advanced driver training could be to introduce a system of positive points, which would be awarded to those who successfully complete approved training schemes beyond the standard driving test. The positive points would help offset any penalties incurred for minor offences, and more positive points could be earned throughout a driver's career for undertaking further training or retesting.
N.8 The most important action that government needs to take, however, is to abandon publicity campaigns based on misleading slogans such as "Speed Kills", which have caused so much damage to Britain's road safety culture during the last 15 years. Emphasis needs to be placed instead on the need for all road users to take responsibility for their own safety, and to accept that they cannot rely on others - or the law - to save them from their own mistakes.
N.9 Everyone needs to be made aware that learning to use the roads safely is a vital life skill, not an optional extra. They should be encouraged to develop that skill to the best of their ability, so as to minimise the risks they pose to themselves and others. Above all, they need to realise that to be safe on the roads requires their full attention at all times.
N.10 This should not apply just to drivers but also to pedestrians and cyclists, who have been encouraged to think that it is entirely drivers' responsibility to avoid them. Over 80 per cent of pedestrian fatalities are precipitated by the pedestrian entering the road without due care48. Some of those accidents could no doubt have been avoided if drivers had been paying more attention or driving more defensively, but the fact remains that all road users have a responsibility towards themselves and others when using the roads.
N.11 Instilling the right attitudes concerning road safety needs to start as early as possible, preferably in schools by increasing the road safety content of the Citizenship curriculum. But it cannot be emphasised too much that this must not mean brainwashing people, whether adults or children, into slavishly following simplistic rules; the road environment is too complex for that, and people have to think for themselves within a framework of basic principles. Education should focus on teaching those principles and the correct approach to individual risk management on the roads.
N.12 Road safety policies based on hectoring drivers and punishing them for breaking inflexible and often arbitrary rules cause great resentment. No one likes being patronised, or being treated like a naughty child, and the result is often the opposite of that intended. The driving public should be treated like responsible adults, which most of them are. Education campaigns should explain the real issues and not try to simplify them into misleading sound bites. Only then can we hope to see any real improvement in road safety in Britain.
Annex O: Reducing deaths and injuries from drink-driving
O.1 Over the last thirty years there has been a welcome reduction in the social acceptability of drinking and driving. This has led to a significant decrease in the numbers of people killed each year in drink-related accidents, from around 1,790 in 197949 to 460 in 199850. Since then, however, deaths in drink-related accidents have risen to an estimated 540 in 200650, similar to the number in the mid 1990s49.
O.2 The reversal of the downward trend has coincided with the reduction in police traffic patrols following the switch to camera enforcement of speed limits. This has led to a perception that there is less likelihood of being stopped than before, so drivers may be tempted to take the risk of drinking and driving.
O.3 The increase in drink-related casualties in recent years has led to a call, once again, for the legally permitted blood-alcohol concentration (BAC) to be reduced from the current 80mg per 100ml of blood to 50mg or even less. Yet there is neither evidence nor logic to support the view that such a reduction would have a beneficial effect. Latest figures show that, in 2005, 24 per cent of adults killed in road accidents where BAC levels were known had more than 80mg per 100ml BAC. Just 3 per cent had a BAC in the 50 - 80mg range51 (and it should not be assumed that the presence of alcohol automatically means that the effect of the alcohol caused the accident).
O.4 The current BAC limit was set in 1967 based largely on the results of the large-scale 'Grand Rapids' study in the USA52. This showed that the chances of being involved in an accident only starts to climb significantly when BAC levels exceed 100mg, so 80mg was chosen as the legal limit to give a buffer of safety. While no fixed limit can be ideal under all possible circumstances, the fact that the current figure has some scientific justification has undoubtedly contributed to its general public acceptance as reasonable. Drivers caught at higher BAC levels are usually afforded little sympathy.
O.5 The recent increase in fatalities in drink-related accidents is the result of more drivers exceeding the current BAC limit - and more pedestrians drinking in excess of the limit for drivers. If drivers are drinking above the current limit, it is illogical to believe that they would observe a lower limit. Furthermore, if a lower limit were introduced and drivers were faced with 12-month driving bans for exceeding it by a small amount, when there was little evidence that their ability was impaired, such injustices could erode public support for the law. This is particularly relevant to the 'morning after' issue.
O.6 There is evidence that, for a given BAC level, an individual is less impaired when the alcohol concentration is falling than when it is rising53. Thus a driver who is slightly over the limit immediately after an evening's drinking is likely to be more of a hazard than one who avoids drinking and driving but still has alcohol in his or her system the next morning. A driver in the latter category may feel - and be - safe to drive, but could unwittingly be breaking the law.
O.7 Some drivers are already being caught with residual BAC levels above the legal limit, many hours after their last drink. Many more people could find themselves in the same position if the BAC limit were lowered. The ABD supports the advice that people should ideally not drink and drive at all, but it would be unjust if responsible individuals follow that advice and then find themselves banned from driving for being a few milligrams over a 50mg limit the next day, when they pose little or no danger to others. Support for drink-driving legislation could then be undermined.
O.8 The ABD does not support a lower BAC limit, therefore, but believes that the existing limit needs to be better enforced. This does not mean that the police should engage in the random testing of large numbers of drivers. They already have the effective power to stop any driver at any time, and they are quite capable of deciding which drivers are likely to be impaired by alcohol. But they cannot exercise those powers if they are not patrolling the roads.
O.9 As proposed earlier concerning the enforcement of speed limits (para. H.8), more resources and greater priority should be given to providing police traffic patrols. Those patrols should use their enforcement powers for the purpose of improving road safety and apprehending serious and wilful offenders, rather than seeking to prosecute large numbers of drivers for the sake of it.
1. Swiss report on road accident trends (2005/2006). Table 4: Road fatalities per 100,000 inhabitants in OECD countries, 1980-2004.
2. Department for Transport, Transport Statistics Great Britain 2005, Table 7.6.
3. Road User Alliance Fact File 2006.
4. DETR, Tomorrow's roads - safer for everyone.
5. Department for Transport, Transport Statistics Great Britain 2005, Table 8.1.
6. British Medical Journal, Changes in safety on England's roads: analysis of hospital statistics, 2006.
7. Department for Transport, Road accident casualties: a comparison of STATS19 data with Hospital Episode Statistics, 2006.
8. J J Leeming, Road Accidents: prevent or punish? Cassell, 1969. (Reprinted 2007 and available from www.abd.org.uk.)
9. Road Safety Act 2006, Chapter 49, section 20.
10. Transport Research Laboratory, TRL Report 548, Vehicle-activated signs - a large scale evaluation.
11. Safe Speed.
12. U.S. Department of Transportation Federal Highway Administration, Synthesis of safety research related to speed and speed limits.
13. Establishing Speed Limits - A Case of Majority Rule. Arizona Department of Transportation. 1999.
14. Evidence contained in Annex E of Department of Transport Circular 1/80, Setting local speed limits.
15. Department for Transport, Road Casualties GB 2005, Table 3.
16. Department for Transport, Road Casualties GB 2005, Table 6c.
17. Transport Research Laboratory, Project Report PR58, Speed, speed limits and accidents, 1994.
18. Transport Research Laboratory, TRL Report 421, The effects of drivers' speeds on the frequency of road accidents, 2000.
19. Transport Research Laboratory, TRL Report 511, The relationship between speed and accidents on rural single-carriageway roads, 2002.
20. Transport Research Laboratory, Published Project Report PPR026, Accident analysis on rural roads - a technical guide, 2004.
21. Department for Transport, Circular Roads 01/2006, The setting of local speed limits.
22. West Virginia Department of Transportation, Safety tips: speed limits.
23. Transport Research Laboratory, Project Report PR58, Speed, speed limits and accidents, 1994.
24. Cato Institute, Policy Analysis No. 346, Speed Doesn't Kill: the repeal of the 55mph speed limit, Stephen Moore, 1999.
25. The Policies Studies Organization, Review of Policy Research Volume 22 No. 4, The evolution and devolution of speed limit law and the effect on fatality rates, 2005.
26. Information from National Motorists' Association of America.
27. Montana Department of Transportation.
28. Department for Transport, contributory factors to road accidents, 2006.
29. Source; Road Traffic Regulation Act 1984, Chapter 27, section 89.
30. Road Traffic Act 1991, Chapter 40, section 40.
31. Road Traffic Act 1988, Chapter 52, section 172, as amended by the Road Traffic Act 1991, Chapter 40, section 21.
32. PA Consulting Group, The national safety camera programme: Four-year evaluation report, 2005.
33. Department for Transport, Handbook of Rules and Guidance for the National Safety Camera Programme for England and Wales for 2006/07.
34. PA Consulting Group, The national safety camera programme: Four-year evaluation report, 2005, Appendix H.
35. Department for Transport, Transport Statistics Bulletin: vehicle speeds in Great Britain 2005.
36. Transport Research Laboratory, Published Project Report PPR173, Development of a speed limit strategy for the Highways Agency - proposed strategy.
37. Oliver Carsten, Institute for Transport Studies, University of Leeds, "Can't Go, Won't Go", 1999.
38. Institute of Advanced Motorists, Advanced Driving Magazine, article on research by Mark Young, University of Southampton, 2006.
39. Al Gullon, AMPS (Absent Minded Professor Syndrome) and 'Black Boxes', 2006.
40. Evening Standard, 27 January 2003, '999 patients "killed by speed bumps".'
41. Transport Research Laboratory, TRL Report 564, Road design measures to reduce drivers' speed via 'psychological' process: A literature review.
42. Transport Research Laboratory, TRL Report 641, 'Psychological' traffic calming.
43. Spiegel Online, 16 November 2006, 'European cities do away with traffic signs.'
44. Paper to Historic Design Champions National Conference, 12 July 2006, by Deputy Leader of RB Kensington and Chelsea.
45. Transport and Road Research Laboratory, TRRL Report LR 499, A study of accident rates amongst motorists who passed and failed an advanced driving test, 1972.
46. Fleet News, report from DSM UK to an At Work Road Safety conference, 2001.
47. General Accident Plc, The accident liabilities of advanced drivers, 1991.
48. Transport Research Laboratory, TRL Report 323, A new system for recording contributory factors in road accidents, 1998.
49. Alcohol Alert, 30 years of the breath test, 1997.
50. Department for Transport, Road Casualties GB 2006, p.26.
51. Transport Research Laboratory leaflet LF2101, Blood alcohol levels in road accident fatalities for 2005 in Great Britain, 2007.
52. R F Borkenstein et al, The role of the drinking driver in traffic accidents (Bloomington, Indiana University, Department of Police Administration), 1964.
53. Kinney & Leaton, Loosening the Grip: handbook of alcohol information.
The Transport Committee report was issued in October 2008:
Ending the Scandal of Complacency: Road Safety beyond 2010 [pdf] The emotional title is a bad start …
The ABD issued press release 641 strongly criticizing the report on 29th October.