1.  This reply is only about laws on speed and their enforcement, not about other traffic regulation. One way of enforcing the law on speed-related offences would be to impose higher penalties. This might be a very effective deterrent and could also help to assuage the justified sense of unfair, inadequate and dismissive treatment that many victims and their relatives feel. This note does not discuss penalties because I have not studied them; no doubt the Committee will be receiving many submissions about them.

  2.  Speed is much more important than is usually realised. It is not only an issue in road safety, although that is how it is often regarded, but also affects fuel consumption, air pollution and noise. Safety is not only about the crashes and casualties that occur but also about the intimidating effect that danger on the road has on people's lives. Travel patterns are distorted, children's play and other street life are suppressed, and, if the travel or other activities do still take place, they are a source of anxiety both to the participants themselves and to others on their behalf. Speed is a much more important factor in causing crashes and casualties than is often realised and the crashes and casualties themselves are both under-reported and undervalued.

  3.  Lower speeds would not only reduce rates per vehicle mile of crashes, casualties, fuel consumption, and emissions of noise and fumes, but would also reduce vehicle mileage, so bringing these costs down further and reducing congestion as well. In towns, the most important way that lower speeds would reduce vehicle mileage would be by removing, or greatly reducing, the deterrent to walking and cycling, but in larger towns they would also help to tilt the balance towards public transport and away from cars and to reduce the length of car journeys. Outside towns, the journey-shortening effect and the influence on modal choice could be very significant, probably much more so than road pricing unless the charges were much higher than is now contemplated.

  4.  Speed limits should be lower on all classes of road. This can be shown even by an analysis that takes account of only some of the benefits of lower speeds and exaggerates the likely time penalties. The case for reducing the default urban speed limit from 30mph to 20mph is very strong. The motorway speed limit should probably be no higher than 55mph, as it used to be in the USA, the limit on dual carriageways perhaps 50, and on most rural single carriageway roads no higher than 40mph (1).

  5.  The best way to enforce speed limits is through the vehicle (see below). For the time being, however, enforcement has to rely on existing methods of police presence, traffic calming and speed cameras. Many more resources should be devoted to these measures, since even on an incomplete cost-benefit analysis they show extraordinarily good value for money. For example, it can be calculated from the study on the cost benefit analysis of speed cameras done for the Home Office Police Research Group that over a 10-year period the NPV/C ratio for the 420 sites examined was 7.6 (2). No major road or rail scheme produces ratios of anything like this size. The analysis took account of the costs incurred by the police and the judicial system in administration, collecting fines, etc. It did not take account of the savings in police costs that would result from the smoother flow of traffic brought about by reduced speeds and fewer crashes. This benefit could, however, be very significant, as could the time savings to occupants of motor vehicles (3). It is also significant that the study was based on the simplifying assumption that slight, serious and fatal crashes were all reduced in the same proportions. In fact, speed reductions always produce higher proportionate reductions in the more severe crashes, to which much higher monetary values are assigned. If this had been taken into account, the benefits in saved crashes might have been more than doubled (4).

  6.  Speed humps and other vertical deflection also shows very high benefit/cost ratios, but they do have some significant disadvantages not taken into account in the calculations. They are not selective, but slow down vehicles, such as ambulances, which ought not to be slowed down as well as those that should be. They can cause an uncomfortable ride to vehicle occupants even when the vehicle is moving slowly. They can create noise and, if drivers speed up between the humps and brake before reaching them, can also increase emissions. They can also give rise to structural damage, for example to the cellars of old houses close to the road.

  7.  These facts strongly suggest that in situations such as 20-mph zones where humps or other means of vertical deflection are now used, speed cameras should often be used instead. Another reason is that whereas the humps have to be paid for from the public purse, speed cameras are in effect self-financing, since the revenue from fines more than covers the costs. At present, however, cameras cannot be used on roads with speed limits of less than 30mph; this restriction should be lifted. The absurd rule that cameras should be highly visible should also be scrapped. Drivers should know that there are cameras in the area in which they are driving, but they should not know the exact sites. My understanding, which should be checked, however, is that neither of these reforms would require primary legislation.

  8.  Top-speed limiters on lorries and coaches have been very successful. We now need variable speed limiters on vehicles of all classes. Speed limiters can either be operated by the driver or activated externally. Drivers would be legally obliged to set the driver-operated limiter at a point corresponding to or below the speed limit on the road on which they were driving. There would need to be some external sign to show other road users and the police where the limiter had been set. This could take the form of a colour-coded light on the windscreen, or perhaps a row of lights, such that if one was on it would indicate that the limiter was set at or below a given speed, if two were on it would mean that the limiter was set within the next higher band, and so on. Better still, it might be possible to use lights to display the exact speed at which the limiter was set. If a driver had some good reason to break the speed limit, for example if a car was being temporarily used as an emergency vehicle, the lights could be made to flash as a warning to other road users. The externally activated limiter involves a GPS receiver inside the vehicle to show where on the road network it is, combined with a computerised map in the vehicle to show the speed limit at that location.

  9.  The externally activated limiter has the advantage that if everything works well it is completely self-enforcing, whereas the driver-operated limiter would still require some further means of enforcement. How big an advantage that is is difficult to say, especially as in practice it would probably be necessary to provide the externally activated limiter with some driver-operated override facility. All the other considerations favour the driver-operated type. The technology is proven, being almost the same as that for cruise control, which has been established for years. The cost would be less, and, above all, it could be introduced immediately, whereas it will be many years before the externally activated limiter could be in general use.

  10.  Despite these facts, all the Government's and the EU's research efforts are going into the externally activated limiter. No reason has been given for the neglect of the driver-operated type. It may be a love of high tech for its own sake, but I suspect that a more important reason is political cowardice. The Government wants to be able to say that it is pursuing the idea of variable speed limiters, but does not want to do anything on the ground now for fear of unpopularity. I hope the Committee will press the Government on this point. Unless the introduction of the externally activated speed limiter can be very much speeded up, the driver-operated type should be pursued.

  11.  An irony is that there is considerable evidence that effective speed control would be popular. Ministers should not listen too much to the vociferous objections of motoring organisations. Most people who belong to these organisations do because they value the breakdown and other services offered. This does not give motoring organisations the right to speak for their members on matters of transport policy. (The Environmental Transport Association and the Association of British Drivers are exceptions. People who choose to belong to those organisations can be taken to subscribe to their views on wider issues.) Governments should undertake their own attitude research. If it did indeed show misunderstanding or hostility on the part of substantial numbers of people, that would indicate a need for education and publicity; it would not be a reason for failing to adopt policies overwhelmingly supported by objective analysis.

  12.  There is something absurd about designing very fast vehicles and then trying to find ways of stopping people from using that speed. Vehicles should be designed with a top speed to match the national speed limit and with much more modest acceleration than at present. This subject is best pursued in the context of the Committee's forthcoming inquiry on Cars of the Future.

REFERENCES

  1.  Stephen Plowden and Mayer Hillman Speed Control and Transport Policy, PSI, 1996, Chapters 9 and 10.

  2.  Andrew Hooke, Jim Knox, David Portas Cost Benefit Analysis of Traffic Light and Speed Cameras, Police Research Series Paper 20, Home Office Police Research Group, 1996. Table 16 shows initial fixed costs (purchase, installation and some other items) of £5,264,000; annual recurrent costs of £3,595,000; annual benefits of £36,969,000 made up of £30,239,000 crash reduction benefits and £6,730,000 income from fines. It was assumed that the fixed costs occurred at the beginning of the first year and the annual cost and benefits at the end of each year. With a discount rate of 6%, the total discounted costs over ten years come to £31,724,000 and the total discounted benefits to £272,094,000.

  3.   Ibid, pages 23 and 27.

  4.  The annual number of crashes saved at 241 sites with traffic light cameras was 115.68. An average value of £57,598 (1995 prices) was applied to this number to give a total annual value of crashes saved of £6,662,937. However, it was estimated (precise data were not available for each site) that 12 of the crashes saved were fatal ones. The value of a fatal crash saved, at 1995 prices, was £945,100, so these 12 crashes alone were worth £11,341,200. The ratio of serious to slight crashes saved was also higher than implied by the figure of £57,598. Ibid, pages 26 to 28.

October 2003