Tuesday, 1 June 2010

Road Safety - not as clear-cut...?

I love counter-intuitive findings. This one has been around for a while though, with only the more thoughtful people realising it:
The trouble with cars these days is that they’re too safe.
Too safe? Yes, indeed.
We need more dangerous cars. A spear mounted on the steering wheel, pointing at the driver’s heart, would do nicely. Cheese wire instead of seat belts would work too. Of course, these innovations would skewer and slice the typical crash-test dummy, but drivers aren’t crash-test dummies. Give them the right incentive and they will drive more carefully, to the benefit of the cyclists and pedestrians.
Do read the whole article; Harford makes the point which I have been trying to make for years, that road safety is a balance of conflicting interests. Safer cars can mean fewer fatalities, or it can mean cars go faster for the same number of fatalities, both of which are good (albeit in rather different ways).

And for those who think that road safety is a simple issue and that the obvious measure will obviously help in the obvious way? Well, consider the findings of:
Sam Peltzman, who in 1975 published a paper demonstrating that drivers did indeed drive more dangerously after mandatory seatbelt laws were passed in the US. He argued that despite technological evidence showing that seatbelts save lives in a given accident, there was no evidence that the seatbelt laws had reduced driver fatalities. In other words, drivers take advantage of seatbelts to drive more dangerously rather than to live longer. More compellingly, Peltzman detected a rise in pedestrian and cyclist fatalities when seatbelt laws were passed.


  1. This is true, but not new.

    I thought it long been established that people need a given level of risk in their lives. Doesnt need to be rock climbing, it can be driving quickly, unsafe sex, causing arguments, getting into fights etc. The preferred level varies from one individual to another but we all need to feel like we're in danger some of the time.

    So making a risky activity that we undertake (eg driving) safer allows/encourages us to do something new or different to restore our risk levels to where we prefer it.

    When you work it out, taking into account higher numbers of cars, increased mileage driven etc the fatality figures haven't changed much for years. We soak up the safety improvements and counteract them with other riskier actions, eg speeding.

    So the spike on the steering wheel would definitely reduce speeding, but some folk would get killed from it as their risk level took them one step too far.

  2. Am I the only one who has noticed, that all councils seem to be reducing speed limits lately, 50 to 40 and 40 to 30. Presumably they are being offered "incentives" by government as with the useless cycle lanes!
    The latest is to reduce the 50 limit to 40 on the A40 near me. On the grounds that there have been two fatal accidents on this section in the last few years.
    But a speed limit would have made no difference. The first was a Fire Service vehicle on its way to an accident, and the second was a night shift worker returning home at about 6am on his motorbike who was estimated to have been doing around 100mph when he hit a cleaning lady crossing the road on her way to work.
    What difference would a speed limit have made to either accident?
    But no doubt it looks good on paper.

  3. The problem with all of these things is that they fail to take account of the human aspect of our behaviour. Last year when Pope Benedict spoke about condoms and HIV/AIDS in Africa, the liberal Western media went for him.

    But as Edward C Green , from the Harvard AIDS research centre, pointed out "in truth, current empirical evidence supports him." He went on:

    In 2003, Norman Hearst and Sanny Chen of the University of California conducted a condom effectiveness study for the United Nations' AIDS program and found no evidence of condoms working as a primary HIV-prevention measure in Africa. UNAIDS quietly disowned the study.

    The problem is the same as you point out here: "risk compensation". He cites a study which found that people who had participated in aggressive condom promotion scheme engaged in more risky sex

    The net result was an increase in sexual risk in the intervention group, as “gains in condom use seem to have been offset by increases in the number of sex partners.”

    This might explain why it is the people in sub-Saharan Africa who have never used a condom have the lowest rates of HIV infection and those who used a condom at last sex have the highest rates (it might also explain why HIV infection in Sub Saharan Africa is in general lower in Catholic countries than in non-Catholic countries).

    As Green said: "Surely it's time to start providing more evidence-based AIDS prevention in Africa."

    Of course, the media didn't really focus on the evidence: the careers of those who have been promoting condoms for the last 30 years are at stake, and if you can protect them while bashing the Pope, so much the better.

  4. English Pensioner: No, you aren't the only one. The 'old' A13 changed from 60mph to 40mph last year, for absolutely no good reason that I could see...

  5. So what is the role of the insurance companies in all this? Have insurance premiums come down? How much work do they save the police?

    Why do 3 points on your licence, ie a speeding ticket, allow them automatically to justify raising your annual premium by 10%? Money for old rope, except (as Mr m reminds) where I'm concerned. If we do have to drive with snails, I'd like to see some positive feedback before resentment sets in.

  6. There does seem to be a general problem with groups that are keen to show that their policies are supported by scientific research, but feel that the way to do this is to suppress the research that gives the "wrong" answer.

    I can't help but think of the tale of Semmelweis broadcast by Stephen Fry on Radio 4 in their "Moments of Genius" series.

    M - you can find many insurers who will take no notice of a single speeding ticket issued by a "safety" camera or van. This says something about the relative safety of the drivers caught by these means.

  7. P,
    Quite, and the pressure groups tend to have in common the wrong-headed assumption that technology is the solution to what are essentially human and personal problems. That's when you get stupid (and in some cases, dangerous) policies. So I think the problem is that as usual, our society doesn't really know what human beings are and what makes them tick and as a result is unable to assess evidence properly.

  8. Personal responsibility, what's that?

  9. It's something we used to use, Blue. It had to be dropped as there was no budget for it.