The safer things get, the more we want to risk

You wear a seatbelt, but you smoke. Taking chances, it seems, is only human

Think about risk. This week, on the day of the Channel Tunnel fire, there was also the first substantial snowfall of the winter. No one, mercifully, died in the fire, but at least two people died on the roads as a direct result of the weather. Yet a fair number of people will be discouraged from using the tunnel, while hardly anyone will stop using the roads.

We have the strangest attitudes to risk. We live in a society which seeks to reduce risk, yet individually we are prepared to take them, maybe even seek them out. Thus we worry about nuclear power and genetic engineering but are perfectly content to drive cars, ride bicycles, or smoke cigarettes.

In a New Scientist study, nuclear power ranked right at the top of people's perceived risks, far above terrorism, Aids and mugging.

Strange? Well perhaps not so strange. There may be a great gulf between the actual dangers of something terrible happening and our perceived fear of it, but there is a certain rationale to our attitudes. For example, we worry much less about the risks we feel we can control, like accidents in the home or when skiing, than those we cannot, like the damage to the ozone layer - or being stuck in a fire under the Channel.

The older we get the more worried we become about risks, because teenagers are notorious for believing they are indestructible. And we discount risks into the future in the sense that we worry far more about things that might go wrong now than those that might go wrong 20 or 30 years hence; if we did not discount risk in this way hardly anyone would smoke another cigarette.

The fact that we are bad at assessing risk has been widely noted. Last year John Adams's controversial book Risk argued that people were prepared to accept a certain level of risk in their lives and when legislation reduced that risk, they found other ways of increasing it. His most cited example was the introduction of seatbelts in cars, which he argued had little or no effect on the road death toll, which was falling anyway. They may have saved the lives of some people inside the cars, but because they encouraged people to drive faster, more people outside the cars were killed.

Another example was the Davy safety lamp in mines in the last century. This operated at a temperature below the ignition point of methane, and because it cut the risk of explosions was credited with saving the lives of thousands of coal miners. Yet it seems because its invention encouraged mining in more dangerous seams, the number of explosions and deaths actually rose in the years after its adoption.

The fact that most people are bad at assessing risk leads inevitably to inappropriate legislation. Politicians naturally and properly have to represent their voters' views, even if this leads to odd outcomes. In the United States, concern about passive smoking has become so intense that in a friend's office in Washington they are not allowed to smoke outside the building on the pavement - they have to move at least 30 feet away from the front office door.

In any case, badly framed safety legislation can be a very inefficient way of improving safety: it encourages service providers to design to fit the law, rather than to make whatever they are doing really safer for customers.

The fact that a lot of safety legislation is misguided, because of a combination of misinformed popular pressure and incorrect assessment of risk, has already received a lot of attention. But there is still an enormous groundswell of pressure to reduce risk and I guess that for another generation at least western societies will continue to regulate in order to try to reduce this risk. Some of that regulation will be genuinely helpful, but a lot won't.

Indeed the more we try to "manage" risk, the more we may increase it. This is one of the points made in a new book on the subject, out this week, by the American economist Peter Bernstein, Against the Gods: the remarkable story of risk. Bernstein argues that schemes designed to cut the risk of a stockmarket fall in 1987 actually contributed to the scale of the crash in October that year.

That is one problem, because when people find that risk cannot be eliminated, they will be deeply resentful. But I think there is a bigger problem which worries me even more. How will society cater for people who want more risk, not less?

Silly question? Absolutely not. Some people do want more risk. Why do Britons, unlike Germans, Scandinavians or Japanese, skip across the road when the pedestrian lights are red? Why has about one-third of the teenage population of the UK taken ecstasy? Why do people smoke?

One of the great challenges I think that developed countries face is to find ways of catering for people's desire to take risks, without getting out of control and damaging the rest of society. You could say that encouraging activities which seem dangerous but actually are fairly safe, like hang- gliding or parachuting, will provide an outlet for this desire. But while that is fine for the organised and relatively well-off, it does not meet the desires of the less organised. Crime is an obvious outlet: for very young offenders there is some evidence that the main attraction is not the money but the buzz.

To say that is absolutely not to argue for laxer safety standards on the Channel Tunnel. That would be absurd. But it is to say that each new bit of restrictive legislation ought to be tested against this criterion: does it take the fun out of life, particularly for young people? And if the answer is yes, then we should think hard before pressing ahead.