Final open letter to Executive Director of PACTS

(If experiencing problems with IE7, please try Firefox, Opera or Safari)

To Robert Gifford

Executive Director, The Parliamentary Advisory Council for Transport Safety

Dear Rob

Failing a response, I propose to draw this one-sided correspondence to a close. I believe that I have established:

a. that the claim that seat belts have saved 60000 lives since the implementation of the seat belt law is nonsense, and

b. that it is your (PACTS’) view that a measure that could be demonstrated to save the lives of motorists would be OK so long as the number of vulnerable road users killed as a result is smaller. I pressed you on this point in open letters 2 and 3 and you have not, as yet, denied it.

When you sent me the paper by Richards et al that I discussed in my third open letter you said the paper “highlights the difference between the number of lives saved overall through the use of seatbelts and the number saved through the implementation of legislation in 1983.” I take it from this that you accept these “highlights” as validation of the claim still on your website that seat belts have saved 60000 lives since 1983.

In my third open letter I said that with such a sudden large increase in the wearing of seat belts (140%) one should see a sudden large downward step in the established downward trend. I asked “Where is it?” I have found it! I was looking in the wrong place.

In the graph below I have reconstructed the analysis of Richards et al that highlights the difference between the number of lives “saved” overall through the use of seatbelts and the number “saved” through the implementation of legislation in 1983.


To accept the claim in the paper by Richards et al that seat belts saved 2500 lives in 1983 you have to believe that 1869 extra lives were saved by seat belts compared to the year before. Since you cannot see such a dramatic effect in the graph of actual fatalities you are forced to believe that in 1983, coinciding with the implementation of the law, there was some dramatic increase in danger on the roads that, but for the seat belt law, would have killed 1869 more car occupants. Even people who can believe six impossible things before breakfast would find that a bit of a stretch.

Why am I doing this? Why am I trying to resolve this issue via open letters on my website? I have no personal desire to cause embarrassment, but earlier attempts to sort this out via personal emails have run into the sand.

In a parliamentary debate in 1979 William Rodgers, then Secretary of State for Transport, claimed “On the best available evidence of accidents in this country – evidence which has not been seriously contested – compulsion could save up to 1000 lives a year” (Hansard, March 22, 1979). This clearly did not happen, but the myth has grown and PACTS has been complicit in this growth.

Consider the following more recent parliamentary exchange in a debate about compulsory cycle helmets:

Mr. Kevan Jones (North Durham) (Lab): There are some who would accuse my hon. Friend of extending the nanny state. Does he agree that the same arguments were used in the 1960s and 70s against the wearing of seat belts, and that the legislation passed in that respect has reduced the number of deaths and the personal tragedy experienced by families whose members would otherwise have died on the roads?

Mr. Martlew: My hon. Friend is perfectly right. We have always seen a knee-jerk reaction against such measures, whether on the wearing of seat belts or preventing drink-driving, but, after a while, such things become common sense and we wonder why we did not do them before. (Hansard 23 Apr 2004 : Column 529)

Your website proclaims that PACTS was set up “as part of the fight to get mandatory seatbelt wearing turned into legislation.” The “success” of the seat belt law might fairly be described as PACTS’ foundation myth. It is not a bit of harmless exaggeration.

You have accepted that the law caused an “appreciable” number of extra deaths among pedestrians and cyclists. Further, as we have seen above, the myth is now being invoked to promote another myth about the efficacy of cycle helmet legislation. So profoundly embedded in the parliamentary mind is the “success” of their seat belt law that no one saw reason to question the assertion of Mr Jones. A compulsory helmet law would significantly impair efforts to promote cycling and kill the London cycle hire scheme; no potential user of the scheme would want to put on a sweaty helmet that did not fit, and no scheme has been proposed to provide clean well-fitting helmets for spontaneous users. For a comprehensive over-view of the cycle helmet law debate please visit nohelmetlaw.org.uk.

No comment yet

7 pings

  1. Robert Gifford says:

    Dear John

    I don’t want to comment on the cycle helmets issue. As you may be aware, the PACTS’ position on helmets is that, yes, they can be effective in certain
    circumstances. However, they are not a universal panacea: we also need to reduce the risk posed to cyclists by vehicles through speed reduction
    measures. Voluntary promotion of helmets seems to be the most appropriate way forward.

    On the effectiveness of seatbelts, the savings figures are based on an assumption that otherwise, we would have done nothing. In other words, we would not have undertaken encouragement of voluntary wearing during the 1970s but left it entirely to car occupants to make up their own minds. Your graph therefore has to forget about the red line and concentrate solely on the green and blue ones.

    I appreciate that we are unlikely to reach agreement on this issue.

    Best wishes


  2. Guy Chapman says:

    Mr Gifford’s reply says that it’s based on “an assumption that otherwise, we would have done nothing” and then introduces further hypotheses which embody precisely the same fallacy as the original basis on which the claim was defended.

    The point remains that the figures are hypothetical and based on the assumption that driving behaviour is not influenced by the driver’s perceived safety.

    This further hypothesis appears to me to violate Occam’s Razor. We know from the Munich taxi driver experiment and other sources that drivers’ behaviour changes with perceived safety. What Mr Gifford is saying is that he should be allowed to make any claim for efficacy he likes based on the assumption that seat belts do not cause this behaviour. I still cannot see any evidential basis for that opinion. Unfortunately I think this is looking like an article of faith not an evidenced position.

  3. Colin McKenzie says:

    The weasel words from Mr Gifford have no place in a scientific debate. It turns out that what sounds like “lives saved by the seatbelt law” actually means “lives saved by seatbelts over a timeframe that happens to start when the law was brought in”. Even on this basis, the number quoted has no scientific basis, being built on a raft of optimistic assumptions.

    The evidence fails to prove any sustained reduction in car occupant casualties as a result of the law. Even Mr Gifford accepts that casualties among other road users went up. The iniquity in the law is that it is people who are forced to wear seatbelts that increase danger to others – those who choose to wear seatbelts do not.

    On this track record, any figures used to argue for compulsory cycle helmet use are likely to be both alarming and untrue.

  4. Tristan says:


    I’m a really big fan of yours… and probably one of the youngest at 25. I do find your line of reasoning very compelling.

    But just trying to be clear. What are you saying that the government should do. It’s one think to refute trumped up claims of lives saved.

    However, Helmet laws and seatbelt laws, to me, seem very different primarily because cars kill others. Therefore, the changed risk perception has the result of driving faster, breaking later etc. resulting more deaths to pedestrians and cyclists. If I understand you correctly, you argument that these deaths are more than the much smaller number saved by seatbelts OR that seatbelts change the risk perception to the point that no drivers lives are saved by seat belts.

    With cyclists wearnig helmets there will be no ‘displacement’ factor. If the risk perception changes and cyclists take more risks it unlikely that many drivers or pedestrians will be injured. It’s going to be the cyclists. So it’s a more simple tradeoff. So the questions, I think, are:

    Do helmets cause cyclists to take more risks?
    Does the wearing of helmets, or the rules abotu wearing helmets, result in drivers being more agressive around cyclists?
    Do the above two result in more crashes?
    Do these crashes result in serious injuries/deaths?

    Are accidents with people wearing helmets generally similar or more/less serious than those who don’t?
    If involved in an accident does wearing a helmet reduce the chance of being seriously injured or killed?

    Relatively, would compulsary wearing helmets result in MORE is less deaths?
    Does self selection of wearing helmet (ie not having a law) change this result for better or worse?

    Have I got this right?

    So assuming I have – it seems a really really hard question to answer. I’d love to know how you’d go about doing that (I’ve been involved with driver simulators… that might be ‘an’ angle).

    Or, will it basically come down to a judgement call. If so, the judgement call that a helmet law is good seems to me to just be as justified as a helmet law is bad. There cannot be an objective right and wrong.

    So, John: Do I have it right?

    What do you think is the best option?

    And, as a bit of a tease,

    Do you wear a helmet riding?

    Tristan Cooke
    Brisbane, Australia

    Good questions.

    What should the Government do? Repeal the seat belt law.

    Statistically cycling is very safe – especially when done by large numbers of bare-headed cyclists as in Denmark and the Netherlands.

    Cycle helmets come with a message that is likely to influence perceptions – “Cycling is dangerous.” So dangerous that in Brisbane it is against the law to cycle without one. I confess I am unclear whether this effect might offset the sense of greater protection provided by a helmet. But two things are clear:

    1. Cycle helmets also send a message to motorists. Ian Walker discovered that passing motorists were 23% more likely to come within 1m of the bicycle when a helmet was being worn. (Walker, I. (2007). Drivers overtaking bicyclists: Objective data on the effects of riding position, helmet use, vehicle type and apparent gender. Accident Analysis and Prevention, 39, 417-425.) See also.

    2. Helmet use discourages cycling (especially in hot places like Brisbane) and severely discourages spontaneous cycling; few people will carry a helmet with them on the off chance that they might want to borrow or hire a bicycle. Perhaps you could report on progress in Brisbane on the CityCycle hire scheme. I have yet to hear of a successful scheme in a jurisdiction with compulsory helmets.

    So my “judgment call” is that helmet laws are bad. They discourage cycling and thereby encourage car-dominance, which makes cycling more dangerous. http://www.nohelmetlaw.org.uk/ provides the most comprehensive statistical coverage of the subject that I know.

    And no, I have never worn a helmet. JA

  5. Mayer Hillman says:

    The advice that Rob Gifford offers is incoherent. Look at Adams’ graph and “concentrate”, as Gifford suggests, on the blue and green lines. What should we conclude?

    In 1983 there is a huge jump up from the blue line to the green line. The difference, he argues, represents the number of lives saved by seat belts. But if there had been 2500 lives saved by seat belts in 1983, as he and Richards et al argue, there should have been an enormous step down in the blue line, actual fatalities. But it’s not there! I cannot decide whether Mr. Gifford is confused by the evident contradiction in his attempt to justify the position that, remarkably, he continues to maintain – or whether he is being disingenuous.

    On cycle helmets he (together presumably with PACTS and those who concur with its recommendations) favours promoting voluntary use. This can only succeed if people are persuaded that cycling without a helmet is too dangerous – try telling that to the Dutch and the Danes! If large numbers were encouraged to believe it to be true, it will kill London’s cycle hire scheme. One can only wonder whether he would welcome such an outcome. Helmet wearing and spontaneous cycling are incompatible. People won’t carry a helmet on the off-chance that they may want to hire a bike.

    It seems to me, and I would hope everyone anxious to see the issue of seat belt and cycle helmet wearing resolved once and for all on the basis of evidence and reason, that the logical way ahead is to stop engaging in sparring through emails, comment on websites and in blogs. These media are an inadequate strategy to that end. It is obvious that a conference should be organised where proponents and opponents present their respective cases and that these are then debated and hopefully a sufficient degree of consensus reached as a result of this process.

    The logical bodies to do so are PACTS and RoSPA. Can they be persuaded to take this on? Were they to do so, it would remove the suspicion that their failure to do so reflects a fear that they would lose the argument.

    This is an important area of public policy that needs resolution on the way ahead, taking account of the wide implications of decisions directly or indirectly affecting most road users. Parliament should not have to continue to be heavily dependent on its Advisory Council’s recommendations when these appear to be open to serious challenge.

  6. Duke Maskell says:

    This is the most bizarre–supposedly academic–disagreement I have ever come across. It’s as if the two sides are playing by two quite different sets of rules (or, rather, as if one acknowledges no rules at all).

    Adams, with patient reasonableness, reiterated, demolishes Gifford’s claim that the seat belt law saved a large number of lives in 1983 and invites Gifford either to give up his claim or justify it. Gifford replies by inviting us, amazingly,to concentrate on the evidence that his claim is demolished–the blue and green lines–but as if it were evidence to the contrary. Can the man not read? Can’t he see what the lines mean?

    Hillman–who along with others does his level best to help him–cannot decide whether he is confused or disingenuous. Well, he is certainly very stupid. And it must be, I think, because he cannot now afford to be anything else. He has sworn black was white so publicly, for so long and made such a pretty career out of it that (short of being threatened with burning at the stake) recantation has become for him a moral impossibility.

    Instead, he generously invites Adams and his other tormentors/interrogators to ‘agree to differ’ with him. What is it the Red Queen said? “Off with his head!”

  7. Harry Daly says:

    And does or doesn’t Mr Gifford think it OK for more cyclists and pedestrians to be killed, so long as even more motorists are saved? It doesn’t seem fair that legislation should protect the relatively safe, behaving dangerously, at the expense of the vulnerable they endanger. But Mr Gifford is ‘Executive Director, The Parliamentary Advisory Council for Transport Safety’ so probably knows better. He shouldn’t be so shy about saying what he knows. I once read an article of Dr Adams’s called, satirically, “And how much for your grandmother?” Mr Gifford should do one, straight, on “How many pedestrians for a motorist?”

Leave a Reply

Your email address will not be published.

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>