Putting railways in their place on the risk map

On my way to a train that runs at less the half the speed it did a few weeks ago, I drive past a sign telling me that 61 people have died on Lincolnshire's roads this year so far. Some 449 have been seriously injured. Thus, in our part of the world, are the police illustrating a truth ignored in the aftermath of the Hatfield crash. Cars kill far more people than trains.

On my way to a train that runs at less the half the speed it did a few weeks ago, I drive past a sign telling me that 61 people have died on Lincolnshire's roads this year so far. Some 449 have been seriously injured. Thus, in our part of the world, are the police illustrating a truth ignored in the aftermath of the Hatfield crash. Cars kill far more people than trains.

The chart illustrates the remarkable contradiction between the perception and reality of different means of getting from A to B. Of course, it does not give the whole story. Walking is not inherently dangerous; it's what others do to pedestrians that drives up the numbers. Nor do accident rates based on miles travelled reflect the risks experienced over any particular period, since we cover many more kilometres per hour in trains, cars or planes than on foot. Even so, this pattern is not a fact of life, sent like weather from heaven. It raises real questions about policy.

We spend up to 150 times as much to save a life on the railways as on the roads. Nor is transport is the only field where illogicality seems to rule. Why does the health service spend fortunes on complex operations, when a fraction of that money spent on controlling the spread of infection in hospitals might save many more lives? Why, in general, does the burden of regulation correspond so erratically to actual risks involved? Why doesn't government use those natty little maps of the likelihood and severity of risk, spreading like measles through corporate boardrooms?

Well - government does. Up to a point. But risk management in government is a peculiar science, whose oddities cannot all be put down to daft policy making. The public sector is fundamentally risk-averse, despite being largely protected from the financial consequences of risky decisions. Their tax-raising powers, after all, make it hard for governments actually to go bust. The explanation is that government's bottom line is not money, but public opinion. And public opinion likes certainty, not probability.

Accidents must never happen again. Safety must be 100 per cent. Stable doors must be slammed behind every bolting horse. The public may, of course, suffer a reaction when they pay the price of certainty (no T-bone steaks, slow trains) but it is hard to predict when and where this will happen. Far safer, if you are a civil servant, to gold-plate every European directive than risk public evisceration for a perceived lack of preventative zeal.

Such attitudes stifle innovation. Paradoxically, they may also make the system slow to respond to new risks, because everyone knows that regulations are easier to add than subtract. And they make virtually impossible the solution of such problems as the disposal of high-level nuclear waste, still piling up homeless in the United Kingdom.

Such pressures may explain a tendency to over-regulate, but not the patchiness of the public sector's risk management. This stems partly from what one observer calls "the problem of time and space". If the thousands of people who die from smoking every year were to keel over simultaneously outside the Palace of Westminster, rather than cough their lives away in hospitals and old people's homes, policy would be very different. Ten separate deaths on lonely country roads are ten personal tragedies; a railway accident is a disaster. Public opinion - and hence policy - is focused on disasters.

Now, this is not entirely illogical. The fact that so many people may be killed at one go in a railway accident does make it reasonable to expect more safety mechanisms to be put in place in drivers' cabs than on the handlebars of my bicycle. But this highlights another feature of public attitudes to risk. We are far less risk-averse when it comes to choices we make ourselves than with respect to things we see as done to us by someone else. Paragliding, rock-climbing, cross-country riding competitions - all involve the voluntary taking of obvious risk. Yet (so far at least, thank God) governments have steered clear of detailed legal regulation.

This important principle makes its impact on the overall level of accidental death, particularly amongst two groups - the young and the old. The second chart shows a dramatic fall in these rates since the beginning of the century. The easiest way to get these down further would be to tackle the causes of accidents in the home. To do so, however, would involve driving regulation even deeper into people's private lives, demanding more fire-doors or hand-rails, banning high cupboards, insisting on "safe" cooking methods. The most zealous civil servant must balk at that.

Sadly, however, the boundary is being advanced. It is driving regulation into parts of the informal economy (rent-a-room, Women's Institute sales) to a degree that is both ridiculous and a barrier to entry into certain markets. We do better to rely, as far as possible, on the principle of choice.

This will, of course, mean that we should not expect policy-makers to deliver the same risk ratios across all forms of human activity. But - up to a point - it can also help inform the increasingly confused debate about the public and private sector's abilities to handle issues of personal risk.

A deeply ingrained misunderstanding of "the profit motive" tends to make people ignore the immense amount of risk, human as well as financial, handled effectively by all sorts of commercial businesses. Academics point out that the wider the choice - that is to say, the greater the competition - the more likely markets are to deliver the different levels of risk that different customers may choose to take. This argument, however, can only be taken so far. It relies on consumers being perfectly informed, which in some markets is self-evidently impossible. It also assumes that the risk is taken by the consumers themselves, rather than by those making the product, suffering its pollution, or meeting it head-on along the motorway. So even highly competitive markets may need some framework of regulation.

Nor does this argument deal with the problem of monopolies, such as infrastructure networks. Here we are seeing a depressing revival of the view that these should be publicly owned. The modern version of this is the idea of a "trust", or "not-for-profit", as if these words themselves somehow magically revived the dead hand of the public sector, removed risk, or provided a perfect substitute for commercial disciplines.

In fact, over the past 15 years, we have learnt a great deal about the effective combination of private enterprise and public regulation of parts of our infrastructure that remain natural or near monopolies. To make it work on the railways, however, we need a government that believes in the structure it is regulating.

Mr John Prescott cannot continue to enjoy the luxury of blaming the system he inherited without making changes. If he continues to try to do so, he will continue to undermine both public confidence and the management of the rail network. Public opinion will become more demanding, not less; so will the companies managing our railways. The chickens will - eventually - come back to roost in the Treasury. Investment in the railways can ultimately only be financed by higher fares or higher subsidies. We may well need both. But the successful implementation of a new rail strategy rests on confidence that both regulator and private industry know what they are doing. The Government has to take a lead in creating that.