The trouble with risk

The other day Sue and I were discussing all the food scares we’ve been through in the course of our married life. If we’d stopped eating all the products billed as suspect, both cooking and eating would be boring (which it isn’t) and our children would be food fads (which they aren’t).

Statisticians despair at our ignorance – which is also shrouded in superstition and old wives’ tales. I enjoy the occasional flutter at Salisbury Racecourse, on the national lottery – or when I buy those raffle tickets. But I am not a serious gambler. To see people – be they rich or poor – addicted to gambling is a terrible thing. Did someone forget to tell them about the laws of chance?

Toss a coin and the chances of it landing on head or tail are identical. If you toss it seven times and each time it lands on tails, what are chances of another tail on the eighth throw? Slim? Sufficient to stake your life against it?

No – for chance has no memory. The chance of an eighth tails is still 50:50. That is hard to accept, isn’t it? It runs counter to intuition. It is tempting to make a bigger wager. Don’t!

The coin example is trivial. Think instead of risks to health, life or property. Think of heart attacks, car crashes, smoking, HIV, crime or earthquakes. Let’s face it; we frequently misjudge the nature and scale of risks in everyday life. As David Cox of the Royal Statistical Society has pointed out, there are over 40 million mobile phones in the UK and laboratory studies and accident statistics are clear about the hazard of using mobiles while driving. But the risk to health from hundreds of millions of calls made from mobiles is largely non-existent - yet it cannot be ruled out.

This is precisely the kind of assurance that experts and politicians are called upon to offer. Proving a negative is usually difficult, sometimes impossible and there is often a mismatch between perceived and actual risk. Take crime. Statistics show that the risk of being mugged is almost the inverse of feeling unsafe. People simply won’t accept that, as police and politicians know!

Even when we know the risks we are running, we may not change our behaviour – smoking, rock-climbing, hang-gliding are good examples. We may even try to defeat measures designed to protect us. If you drive a car faster when wearing a seat belt, or ride a bike recklessly when wearing a helmet, you are indulging in ‘risk compensation’.

The one thing we must not do about risk is to ignore it. We must understand it – and make judgments. When something does go wrong in this litigious age, someone must be blamed. Accidents aren’t allowed to happen any more, however long the odds. The compensation culture dictates that such-and-such ‘must never happen again’ – like lightening not striking twice, no doubt. Instead we adopt the precautionary principle. Then, voluntary avoidance becomes a legal ban. There is an economic and social price to be paid for this. Just how dangerous would a tunnel past Stonehenge be? Wiltshire Fire Brigade says it is a fire hazard. How dangerous would wind turbines be next to the labs at Porton Down and so close to Boscombe Down airfield? Yet a Defence Minister announced in the Commons this month that development plans for Porton Down will consider the use of wind turbines.

Many policy decisions on issues from vaccines to GM technology, from transport to nuclear power, from genetics to insurance, depend on understanding risk. These issues are the stuff of politics.

Maybe the three “R’s” of schooldays should become four – reading, writing, ‘rithmetic … and risk.

Robert Key
18th January 2004