House of Commons - Transport

Select Committee on Transport Seventh Report

7 Graduated driver licensing

88. Graduated driver licensing systems provide a staged progression from initial learning to unrestricted solo driving. Restrictions are imposed during early driving to reduce exposure to the most high risk driving situations. Graduated licensing systems are in place in several states of USA, Canada, Australia, New Zealand, Sweden, Norway, Finland, France and Northern Ireland. These systems have different components and impose different restrictions.

89. Examples of elements used in different countries include restrictions on where drivers may drive (the types of road that may be used), when they may drive (night-time curfews), with whom they may drive (restrictions on passengers) and under what conditions (such as permitted alcohol levels). Requirements such as additional training, or periods of crash and conviction free driving, are also intended to encourage safer driving practices. As drivers move through the licensing system the restrictions are lifted until they equal those that apply to the fully licensed driving population.

Perspectives on graduated driver licensing

90. We note that the Department for Transport's written submission stated that in April 2004, following consultation, it had decided against introducing restrictions for newly-qualified drivers. It gave the following justification: "Experience is a matter of development, but it cannot be right for the passing of the test to leave substantive questions about readiness to drive."[137] This argument is severely undermined by the fact that such large numbers of newly-qualified drivers (18%) go on to have crashes in the first year after qualifying. Clearly the driving test is not adequately ensuring the safety of novice drivers.

91. Two days before giving oral evidence, the Minister wrote to the Chairman of the Committee stating that the Government had decided not to pursue restrictions as part of a graduated licensing system, because "we see the reform of driver training and testing as an alternative approach which addresses the underlying causes of the concerns we have with novice drivers."[138] Yet research commissioned by the Department had already concluded that, "There is little research evidence that increased formal driver training improves safety."[139] Whilst we share the Minister's hope that a reformed education and training framework will instil a more responsible approach to driving; there exists little evidence that this will be adequate. It is a fact that novice drivers face a very considerably inflated crash risk when driving late at night, with passengers, and after consuming even small amounts of alcohol. Over-confidence and inexperience result in poor driving choices and ultimately collisions and casualties.

92. When questioned by the Committee in oral evidence, the Minister was slightly more open-minded to the prospect of graduated driving licence systems, stating "I am not going to close down any options at this stage because it would be wrong to do so […] but instinctively I do not see that [graduated licensing] as being the answer […]"[140] When considering the potential of graduated driver licensing systems, we hope that the Government will be led by the evidence, and not by a blind hope that more, and more effective, driver training will be adequate to prevent the large numbers of novice driver casualties.

93. In oral evidence, the Minister also suggested that where graduated systems had been introduced in Europe, these countries had only managed to reach casualty rates already achieved in the UK. This picture is, however, a little misleading. There are almost no restrictions imposed on solo driving by novice drivers under any of these European systems. They tend instead to involve different treatment of violations by novice drivers. By comparison, the graduated licensing systems in New Zealand, USA, Australia and Canada include actual restrictions on solo driving by novice drivers, for example, night restrictions, passenger restrictions and lower alcohol limits. These countries have lower novice driver casualty rates than the UK. For example, an analysis of young driver deaths, as a proportion of all driver deaths, taking into account the population of young people, showed that in Britain, the proportion of young drivers killed per total drivers killed was 2.9 times greater than the proportion of young people in the population. In New Zealand this is 1.9, in Canada 2.2, in Australia 2.2, and in USA 2.4. This suggests that the distortion of risk for young (and novice) drivers is more pronounced in the UK.[141]

94. In 2002, the Department for Transport commissioned a study by TRL into the effectiveness of graduated driver licensing. This study looked at the experience and effectiveness of restrictions in other countries. The report concluded that several elements of licensing systems had been effective in other countries and could be considered for introduction in Britain. The authors acknowledged that there were difficulties in translating experiences from other countries into the UK, and predicting what the effect would be, because the training and testing frameworks, as well as the road environment, were so different. However, the report stated,

"Despite this caveat, a serious case can be made for introducing some elements of graduated licensing, or graduated learning, systems in Britain […] Novice drivers in Britain do have problems with alcohol, night driving, and passengers, and there is good reason to expect benefits from measures that address these problems directly."

95. A more recent study for the Organisation for Economic Co-operation and Development and the European Conference of Ministers of Transport collated the findings of recent international studies to conclude that most evaluations of graduated driver licensing show significant reductions in crashes and fatalities, albeit with great variation. Some studies have reported a reduction in crash rates of 4%, whereas others have reported reductions as large as 60%.[142] The report notes:

143

96. Primarily as a result of this international evidence, there is a very large degree of support for a graduated driver licensing system, from a wide variety of witnesses: safety organisations, academics, researchers, the police, the insurance industry, and driving instructors.[144] Professor McKenna, of Reading University, identified that the tendency to frame restrictions as "deprivations of freedom" meant they are often perceived as controversial and contentious, but there was a wealth of evidence that introducing and enforcing driving restrictions had reduced casualties.[145] He pointed to seatbelt-wearing, motorcycle helmets, speed limits, and drink-driving legislation as examples of effective restrictions in the UK.

97. Although there was strong agreement that graduated driver licensing should be introduced, there was less consensus over the detail of what the restrictions should be.

RESTRICTION ON DRIVING AT NIGHT

98. The research evidence indicates that night-time curfews are one of the most beneficial elements of graduated driver licensing in lowering crash involvement, and severe crashes in particular.[146] In North Carolina, a graduated driver licensing system was introduced with a night-time restriction beginning at 9 p.m. The system resulted in a 43% crash reduction during the night-time restriction period, compared to 20% in the day, suggesting that 23% was due to night-time restrictions alone.[147] Michigan introduced a midnight-5 a.m. restriction, which resulted in a 53% reduction in night-time crashes, compared to 25% reduction in crashes overall. Research undertaken for the Department for Transport concluded that, "There is evidence that night-time restrictions can be very effective at reducing night-time accidents […] though effectiveness will depend on the level of enforcement."[148]

99. The inflated crash risk for novice drivers at night-time is thought to be a result of a combination of factors. Some novice drivers will never have driven during darkness ahead of passing the driving test, despite the fact that driving in darkness requires different skills to daylight driving.[149] Perhaps more importantly, however, night-time driving tends to be for different purposes to daytime driving, for example, it is more often a social and recreational activity. As a result, the likelihood of driving while under the influence of alcohol or drugs or while fatigued is much higher at night. Peer pressure is also likely to be a significant contributor to the driving choices which result in night-time collisions. Driving exposure is itself higher because young novice drivers in particular are more likely to be on the road late at night, compared to the general driving population.[150]

100. The RAC Foundation identified that 50% of accidents involving young male drivers which result in death or serious injury occur at night, compared with 35% for older drivers.[151] The Department for Transport identified that collision numbers are especially high for 17-19 year olds between 9 p.m. and midnight.[152] Norwich Union told us that young drivers account for 45% of road fatalities between 11 p.m. and 6 a.m.; are 10 times more likely to have an accident at night; and are 56% more likely to suffer an injury between 1 a.m. and 5 a.m.[153]

101. This evidence indicates that restrictions on driving without a supervising driver at night-time would be effective in reducing deaths and injuries. Restricting night-time driving raises some concerns in terms of social inclusion and accessibility. The Department has identified that the risk is inflated from as early as 9 p.m. and a curfew which began this early would certainly affect those novice drivers who need to travel by private transport for work in the evenings, and it would be a constraint on the social life of many more.[154]

RESTRICTION ON DRIVING WITH PASSENGERS

102. Driving with passengers has been show to significantly inflate the crash risk of novice drivers, and young male drivers in particular.[155] The crash risk increases with each additional passenger carried, and it is nearly three times higher when carrying three passengers, than when driving alone, as demonstrated in Figure 1.[156] Not only does the chance of a collision occurring increase, but with a car full of passengers, the number of casualties also rises sharply. We heard from Mr Cunliffe, of Lancashire County Council, how the number of passenger casualties in novice driver crashes was a persistent concern.[157] The very real risks associated with young novice drivers carrying passengers mean that young male drivers are now the biggest killer of young women in this country.[158]

Figure 1: The effect of passengers on young driver fatalities

103. The official USA estimates state that night-time driving restrictions combined with peer passenger restrictions are linked to crash reductions of 60% during the hours of restriction.[159] A study of the impact of passenger restrictions on casualties, taking into account different compliance levels, found that with 90% compliance of passenger restrictions, there would be 31-42% fewer road deaths among 16-17 year old group, and with 50% compliance, this would still result in a 15-22% reduction in deaths. The research report commissioned by the Department for Transport concluded that, "Given the association between passengers and accidents, particularly amongst teenage drivers, introducing passenger restrictions for drivers when they first begin driving unsupervised is an option that merits serious consideration."[160]

104. Again, such restrictions would create difficulties for some people, for example those who learn to drive specifically because they need to carry people, including children.[161]

ALCOHOL LIMIT

105. Novice drivers (and inexperienced drinkers) are at increased risk of a collision when driving after consuming alcohol.[162] Studies have focused on young drivers in particular, and research has proven that for a substantial proportion of young drivers, crash risk increases at lower concentrations of alcohol than is the case for older and more experienced drivers.[163] Even small amounts of alcohol have a substantial effect on the driving safety standards of novice drivers.

106. Indeed the European Commission has recommended that an alcohol limit of 0.2 g/l be set for young drivers.[164] In many parts of America, Canada, and Australia, novice drivers already have an alcohol limit of between zero and 0.2 g/l.[165] In the UK, the permitted blood alcohol concentration is 0.8 g/l for all drivers. The introduction of laws in the USA which permitted licence confiscation for drivers aged under 21 with a blood alcohol concentration of up to 0.2 g/l, led to a 21% reduction in single-vehicle, night-time fatal crashes among drivers under 21, compared to states that did not introduce such laws.[166]

107. The OECD report examined the casualty reduction effectiveness of reducing the permitted alcohol level for novice drivers.[167] It found that, starting from a permitted blood alcohol level of 0.8 g/l, it would be effective in casualty reduction terms, to lower this to zero or 0.2 g/l. In contrast, it found that lowering the permitted blood alcohol level from 0.8 g/l to only 0.4 or 0.6 g/l did not produce significant reductions in alcohol-related fatalities.

108. Several witnesses supported a lower permitted alcohol content for novice drivers as part of a graduated driver licensing system.[168] This support was based on the evidence that the driving standards of novice drivers are more readily affected by alcohol, and that drink-driving offences are also more prevalent among young drivers.[169] PACTS noted that breath test statistics show that young men under 21 are persistently around 12% of people found guilty of drink-driving, far disproportionate to their licence numbers.[170] Compared to drivers of all ages, young drivers aged 17-19 are 10 times as likely to have a drink-drive crash and young drivers aged 20-24 are 4.5 times as likely to have a drink-drive crash per mile driven.[171] A small number of witnesses recommended that the permitted blood alcohol limit be lowered for the entire driving population.[172]

109. There was opposition to the proposal from those witnesses who were against graduated driver licensing in general.[173] The opposition was based on a belief that having separate limits would send a confusing message about drink-driving, which risked obscuring the central message of "don't drink and drive". The authors of the 2002 TRL report suggested that a lower drink-drive limit as part of a graduated system might instil safer drink-driving habits even after the restricted period had ended. They also acknowledged, however, that the drink-drive problem was most prevalent among people in their early 20s and so raising the alcohol limit at just the point when drivers might enter this high risk demographic could be problematic.[174] There were particular concerns about the practicality of enforcing two different alcohol limits. We recommend that the Department for Transport, with the Home Office, should evaluate the enforceability of two blood alcohol concentration limits: one for novice drivers, one for the general driving population.

110. Based on the evidence available, we believe there is a case for reducing the permitted blood alcohol concentration from 0.8g/l to zero (or 0.2g/l, which in practice is effectively zero) for novice drivers. If the Department introduces a lower permitted blood alcohol concentration for novice drivers it must be assiduous in countering any impression that it is acceptable for more experienced drivers to drive under the influence of alcohol. We understand that the Department is to shortly consult on proposals to address the problem of drink-drive collisions. We welcome this much-needed investigation and look forward to a thorough examination of what should be the permitted blood alcohol concentration for drivers. The Department for Transport must tackle drink-driving through ongoing publicity and enforcement campaigns targeted at all drivers, including the young and newly-qualified.

LIMITING ENGINE SIZE

111. A small number of witnesses suggested that there should be a restriction on the engine size for novice drivers.[175] Figures from the Department show that novice drivers already tend to drive vehicles with smaller engines than other drivers. The average engine capacity for drivers aged 17-19 was 1366 cc; for drivers aged 17-25 it was 1479 cc, and for drivers aged 26 and above it was 1758 cc.[176]

112. There is little research evidence to support restrictions on engine size for novice drivers. According to the OECD report, "Only the Australian states of Victoria and New South Wales restrict novice driver access to high power vehicles […] There is considerable doubt about the effectiveness of this measure."[177] This may be due in part to the fact that many of the crashes caused by novice drivers result from travelling at inappropriate speed for the conditions and losing control, rather than driving at very high speeds per se. Cars of any size engine allow novice drivers to travel too fast in residential and urban areas, and too fast for bends on rural roads—which is where many collisions occur.[178] The high insurance charges for novice drivers make powerful vehicles prohibitively expensive for many young and novice drivers.

Proposal for graduated driver licensing

113. Taking into account the international evidence, we recommend that restrictions be introduced for newly-qualified drivers. These restrictions should be framed in a way least likely to impinge on novice drivers when the risks of collision are lower.

Novice drivers should be prohibited from carrying any passengers aged 10-20 years, between the hours of 11 p.m. and 5 a.m.
The permitted blood alcohol concentration should be reduced from 0.8g/l to zero (or 0.2g/l) for novice drivers.

114. Our recommendation is based on the research information already published. There are decisions to be made about how long the restrictions should be in force. Witnesses' suggestions varied from "a few months"[179] to two years.[180] There is also the question of whether restrictions should be lifted after the passage of time, after a second driving test, or on evidence that further training had been undertaken. There was some support for a second driving test.[181] There are also arguments about whether the restrictions should be lifted sooner for novice drivers aged over 25 years (as in the New Zealand system), according to the risks involved. All these decisions about the detail of novice driver restrictions under a graduated driver licensing system should be robustly grounded in the evidence of their casualty prevention potential. As an initial proposal, we suggest that the restrictions for novice drivers should apply for 12 months after passing the test.

115. The Department should undertake more research on exactly what combination of restrictions in a graduated driver licensing system would be most effective in reducing road death and injury among novice drivers, but this must not be an excuse for delaying action. Having implemented restrictions on novice drivers, the Department must closely monitor the impact. It must check for any unintended and undesirable results, and be ready to modify the system if negative safety effects are identified.

ENFORCEMENT OF GRADUATED DRIVER LICENSING

116. One of the main objections to a graduated driver licensing system is the difficulty faced in enforcing such a system.[182] The AA Motoring Trust told us: "We have concerns about any new laws being applied to new drivers unless there is a considerable level of enforcement applied to them. Otherwise new drivers will lose respect for the law."[183] The Minister shared these concerns: "I have not heard any convincing case from anybody yet about how it could be enforced."[184] The Department also emphasised that those novice drivers who were most unlikely to obey traffic laws as a whole, would be very unlikely to comply with such restrictions.[185]

117. Other witnesses suggested that enforcement of a graduated driver licensing system would be no more difficult to enforce than existing traffic law.[186] Professor McKenna advised "There are difficulties in enforcement […] but some of the police simply say that those difficulties are the same difficulties they have in enforcing seatbelts, mobile phones, et cetera."[187]

118. Advocates of graduated driver licensing suggested that the introduction of the restrictions would lead to a cultural change, and that for many people it would become self-enforcing. The role of parental enforcement for young novice drivers was identified.[188] We heard that the experience in the USA had been that the parents were the main enforcers of the restrictions. Restrictions on novice drivers would act as a deterrent to driving in certain high-risk conditions, and where the restrictions were violated, this would be enforced retrospectively by the police.

119. Effective police enforcement would require some changes to the law. For example, Brake advised that in the USA, the experience showed it is difficult to enforce without the power to randomly stop drivers, because it is impossible to tell how long someone has held their driving licence while they are driving along the road. It was suggested that some form of visible indication of novice driver status would be required on the outside of the vehicle, such as 'P' plates.[189] The Police Federation advised that to be enforceable, graduated driver licensing would need to be introduced with a more sophisticated driving licence, which would indicate driver status.[190] The Police Federation also suggested that drivers would have to be legally required to carry their driving licence.[191] RoadPeace suggested that with moves to introduce compulsory ID cards, the requirement to carry a driving licence while driving should not be problematic.[192] The level of roads policing would need to be enhanced if this system was to be enforced properly.

120. We acknowledge that enforcing a graduated driver licensing system may be complex. However, we accept the view of the police and others that it would be no more difficult than enforcing some existing traffic laws. In practice, it is likely to be largely self-enforcing, and where this is not the case, it would be mainly enforced retrospectively. It would provide a deterrent to high-risk behaviour and would give robust support to parents and guardians of young novice drivers, who seek to control their vehicle use.

137 Ev 85 Back

138 Ev 99 Back

139 C Baughan, H Simpson (2002) Graduated driver licensing - a review of some current systems, TRL Report 529 Back

140 Q363 Back

141 Organisation for Economic Co-operation and Development (OECD) / European Conference of Ministers of Transport (ECMT) 2006 Young Drivers: The Road to Safety Transport Research Centre, OECD Publishing, page 28, Table 1.1 Back

142 ibid. See Hedlund et al 2003; Senserrick and Whelan 2003; Hedlund and Compton 2004; Hedlund and Compton 2005; and Hartling et al 2005 Back

143 ibid Back

144 Ev 1, 6, 10, 25, 35, 56, 59, 70, 128, 137, 144, 151, 155, and DIA (2007) Safer drivers, safer roads: Saving young lives through better driver education Back

145 Q270 Back

146 Organisation for Economic Co-operation and Development (OECD) / European Conference of Ministers of Transport (ECMT) 2006 Young Drivers: The Road to Safety Transport Research Centre, OECD Publishing, page 141 referencing Senserrick and Whelan 2003 Back

147 ibid, referencing Foss et al 2001 Back

148 C Baughan, H Simpson (2002) Graduated driver licensing - a review of some current systems, TRL Report 529 Back