mindhenge

Home - politics - risk - article
Home - society - criminal justice - article

A risky business

22 December 2003

Governments cannot legislate against things going wrong, and efforts to reduce the number of failures below a certain level produce negative returns.

The conviction last week of Ian Huntley for the murder of the Soham girls allowed media attention to focus on the startling failure of the vetting procedures that had allowed someone with his history to be appointed as a school caretaker. There seems little doubt that the procedures did fail, and that if they had not failed Huntley would not have been appointed and would not have been in a position to gain the confidence of the two girls. But the press reaction to this failure says something important about public attitudes to risk. These attitudes have a big impact on policy on a wide range of issues.

The problem with risk is that it appears differently in the general and in the particular case. If the death rate from an illness is one in a million, the risk in virtually non-existent in a generalised sense. But if you are that unlucky one person in a million who contracts the illness, the prior risk of doing so will seem to have been frighteningly large and real. It is the failure of the public to distinguish general from particular risk that has sustained the National Lottery. The one-in-ten million or more chance of winning the jackpot is so vanishingly small as to be non-existent as a general proposition. But because, most weeks, somebody does win it, it is open to everybody to project themselves into that particular case and say, as the slogan used to have it, "It could be me!"

Events such as the Soham murders are both common and rare. They are common in the sense that the abduction and murder of a young person is a recurring news item. They are rare in the sense that the absolute number of victims of this sort of crime is exceedingly low. The likelihood of it happening to a given individual is so small that logic might question whether there is any real point in trying to guard against it.

The point here is that precautions also work in a generalised way. The principle is that by identifying whole categories of people for exclusion from jobs having contact with children, the risk of abduction and murder is reduced. It is true that, if everybody within those categories was somebody who would offend if not prevented from doing so, the reduction in offences would be considerable. In reality, however, the majority would not offend, while some offences will be committed by people who are not in one of the categories identified as presenting a risk. The precaution, therefore, has a far less dramatic impact upon the crime statistics than at first might appear.

I recently attended a Neighbourhood Watch meeting at which the police constable in his address gave out two conflicting messages. The first was that crime in the area was exceedingly low - five domestic burglaries in the past year. We ought not, therefore, allow ourselves to feel anxious about crime. The second message was that we should all fit our houses with burglar alarms, since burglars would invariably pass by houses so equipped and move on to the next one.

Now, although it is true that five burglaries is five burglaries, the likelihood of reducing that number by installing burglar alarms is small. The most that can be hoped for is to displace the burglaries to more vulnerable premises, which, in terms of public policy, is no use at all. The significant factor in this is not opportunity as such but the extent of the desire to burgle. In other words, the reason there are so few burglaries is that there are very few would-be burglars on this patch. But because there will always be some, there will, equally, always be some burglaries, no matter how many of us fit alarms to our houses.

The principle applies equally to child-killers. Very few children are killed by strangers, but no matter what precautions are taken there will always be some. When it comes to government policy, therefore, it is futile to plan to prevent all such offences and dangerous to encourage such an expectation. Politicians quickly succumb to temptation in relation to this; by conniving at the public's outrage when things go wrong the government ends up taking responsibility for things it cannot reasonably do. The consequence is skewed decision-making, like the expenditure of billions to reduce an annual death toll of a hundred or two on the railways when the same money could save very many more lives if applied to road safety or directly to the health service.

When politicians respond to what people believe to be the case rather than to what is actually the case, they make bad policy. If people believe that child murder by strangers is a serious social problem, the government can be coerced into spending valuable resources even though it is an area in which it can do little good. That would be fine if resources were infinite, but the reality of government is that money that is mis-spent does the positive harm of denying another area of need that could otherwise be met.

When a man-made disaster occurs, the assumption always is that the disaster should not have happened and something has failed. This approach, however, assumes a risk-free environment, for it is only in such an environment that disasters cannot happen. In the real world of risk disasters do occur, and it is therefore meaningless to say that they should not. Logically, the opposite is the case: where there is risk, disaster should be expected: that people and systems should fail, killers should kill, railway signals should be passed at red. Effective management may reduce them, but it cannot eliminate them entirely.

Policy suffers from the failure to appreciate this. The gap between people's every-day experience (that their children are safe from child killers) and their "worst-case" perception (that their children are not safe because somebody else's proved not to be) needs to be closed if policy-makers are not to be constantly in search of the perfectly safe society at the end of the rainbow. And for this to happen, politicians both in government and opposition must stop feeding bogus public notions about risk. Governments must stop claiming to seek to eliminate it and oppositions must stop talking it up and using it as a stick to beat the government. My policeman needs to be telling me that the crime rate in my area is so low that there is no point in having an alarm fitted.

An exercise in public education is needed to encourage people to see risk for what it is, and to accept that unlikely happenings are precisely that - unlikely. For so long as people continue to believe that they have a realistic chance of winning the lottery jackpot, this task may seem an uphill climb. But politicians would make their own lives much easier in the long run if they could manage it.