Seat Belts: the debate goes on, and on

Letter accepted for publication in Significance, December 2008. This is a much abbreviated version of the letter submitted.

Apologies for my delayed reply to the Controversy piece by Richard Allsop, et al  (Significance, June 2008) – challenging my piece – “Britain’s seatbelt law should be repealed” (Significance June 2007). The myth that seat belt laws save lives is so deeply entrenched that I no longer entertain hopes of their repeal. But I take a tiny bit of consolation from the creeping acknowledgement of “risk compensation” – the idea that people, in this case drivers, respond to changes in their perception of risk.

The debate has shifted from denial of the existence of the phenomenon to an argument about whether in particular circumstances the compensation is partial, complete or more than complete.

Let us for the moment grant Allsop et al their dubious contention of “many more deaths” saved than caused by seat belts.  Who are the saved and who are those sacrificed for their benefit? The saved are people in cars; the lives sacrificed are those of pedestrians and cyclists. The best protected (and usually the economically best off) are provided further protection at the expense of the most vulnerable.

This is a perversion of cost benefit analysis. The central tenet of cost benefit analysis states that a change from the status quo can only be considered an improvement if it makes at least one person better off while leaving no one worse off. Since there is no way of compensating a dead cyclist or pedestrian, their argument fails. Or, in non-economist speak, it is unfair.

For many decades road safety measures have emphasized deference to traffic. Pedestrians are channeled by guardrails or forced to use underpasses and footbridges. Cyclists are offered inadequate cycle paths and encouraged to believe other roads are dangerous. Policy has been to withdraw vulnerable road users from the threat, rather than to withdraw the threat from the vulnerable. The group most seriously affected by this policy is children. The fears of parents and the admonitions of safety campaigners have led to their almost complete withdrawal from the threat. Traditional children’s independence has been lost, and with it a host of experiences vital to their physical and social development.

We can end on a more cheerful and constructive note. Seat belt laws rest on a model of human behaviour that assumes that motorists are stupid, obedient automatons who are unresponsive to perceived changes in risk and who need protecting, by law, from their own stupidity. The idea of risk compensation underpins an alternative model of human behaviour: we are intelligent, vigilant, responsive to evidence of safety and danger and, given the right signals and incentives, considerate. Road users – motorists, pedestrians and cyclists – are now discovering, in pioneering shared space schemes, that safe and attractive urban environments can be devised to encourage the convivial coexistence of all road users.

No comment yet

4 pings

  1. Chris Owen says:

    Please do keep up the debate on this blog as there is a whole new generation of people out here. Your subject is fascinating and of course important.

  2. bent anker Nielsen says:

    I have been arguing your basic point on the influence of the perception of risk to actual behaviour for decades. To no great avail, I regret to say.

    Even accepting the basic premise, people (including politicians) favour “security/safety” (= low percepted risk) to low actual risk in forming their opinions on what needs to be done.
    The argument: “Well I understand, but nobody else will!”

    Kind regards
    Bent Anker Nielsen

    (Chief Speech Writer for the Danish Minister for the Environment – including Climate and Energy – for well over ten years).

  3. Tristan says:

    John and Bent,

    It’s not lost on everybody – some in the medical community are taking notice:


    Tristan Cooke
    PhD Student in Risk and Human Factors

  4. David Goldsmith says:

    I was wondering how air bags might fit in to the discussion on risk compensation. Are they compulsory anywhere? Or so universally fitted as to have the same effect?
    My instinct suggests that the risk compensation effect will be less than that of seat belts or even anti-lock breaking because: either one is completely unaware of their presence or, if one thinks about them, their deployment is rather alarming. Perhaps the negative effects of risk compensation can be reduced by discouraging risk management techniques that are themselves rather benign and comforting – like a seat belt.

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>