Speed cameras and statistical ignorance

A new study shows that government claims about lives saved by speed cameras are overstated. This is ammunition against the free-spending little people who run our local authority highways departments. As recession closes in, councillors and others who have been rubber-stamping big budgets are going to have to start questioning what the money is for and why it is necessary to spend it.

Researchers at Liverpool University have knocked Government claims that 100 lives a year are saved by speed cameras. Whilst speed cameras do reduce accidents, the numbers are exaggerated. The research shows a fall in accidents of 19% compared with the claimed 50%.

Does this matter very much, you might ask. After all, this Government belches out false statistics daily and has, indeed, devoted more energy to rigging the apparent outcomes of initiatives than it has on the initiatives themselves. It does matter, and for reasons which go beyond the actual facts behind this research and beyond motoring. Money is wasted in vast quantities on things which make little difference; things which really do matter are neglected in favour of those which yield apparently good outcomes; the police, who need all the friends they can get at the moment, are tarred with the fall-out of policies to which they do not necessarily subscribe; and any little surviving regard for government (as opposed merely to this Government) takes another pasting.

The Liverpool research recognizes what policemen and highways officers do not – that several factors contribute to an accident, including some which are entirely random. Dr Linda Mountain who led the research put it this way:

“Although some parts of the road network are undoubtedly more dangerous than others, there is also a degree of randomness in where accidents occur – driver error, bad luck etc. – which means that an accident can happen anywhere”.

Mountain used Empirical Bayes to assess the likely effect of random factors on the statistics. This approach to probability theory takes account of known facts and empirical measurements which compare the context being reviewed with other relevant contexts. The approach taken by the researchers allowed them to factor in known and estimated variables and to take account of randomness.

This is all rather different from the approach taken by the police, highways officials and others who make up the speed camera partnerships – or “safety cameras” as they like to call them to give a positive spin on implications of punishment and stealth taxation which their common name implies. The application of Bayesian Probability to road safety is not likely to cut much ice with a Plod sucking his pencil and a county highways officer scratching his bottom in a vain attempt to stimulate thought. Here in Oxfordshire, the expression “thick as a highways officer” is in general use to describe anyone who was at the back of the queue when the brains were handed out, and the police are not renowned for their intellectual skills.

The policeman is motivated by the idea of delegating part of his job to a machine. The simplistic notion that speed equals danger converts easily into the idea that if you can catch people who drive fast – an easily measurable factor – you have solved the problem and can retire to your desk to fill in forms, The biggest single element of randomness – bad driving – is no longer supervised. White van man can tail-gate, overtake on blind corners and do all the other things which cause accidents without fear that he will be caught, as long as he knows where the speed cameras are. The cameras also have the plus of raising revenue without difficult subjective decisions (all that thinking!) and contested court hearings.

The highways officer is driven by a number of factors, none of them actually based on a proper assessment of hazard and risk – these are, in fact, interchangeable terms to people who are this stupid. “Risk” has become the possibility of being blamed for something, and is defined by two things – whether an accident has happened at least once before at a particular place, and the number of people who cry that “something must be done”. A single accident, whatever its cause, is enough if it is followed by lots of shouting.

The best example of this is on Oxford’s Eastern Bypass where a woman caused a serious accident when her over-laden car crossed the central reservation and hit another. The woman was clearly negligent and was jailed for it. Speed per se played no part as a cause of the accident. Yet within days, Oxfordshire County Council agreed to spend £600,000 putting concrete barriers down the middle of the reservation and imposed a speed limit of 50 mph. No consideration of causation, probability, cost or balance against other priorities played any part in informing the decisions. An accident had occurred, people were whining about it, and the officers’ recommendation to do something – anything – was hastily rubber-stamped by the hopeless little people we elect as councillors. The difficulty here is the reluctance of elected councillors to ignore the technical people they employ, however incompetent those people turn out to be.

Highways officers are driven by other things beside fear of being blamed. They love signs; imposing restrictions gives them a power they otherwise lack – look at me, they are saying, I may only be a despised little pen-pusher with a whining voice and a clip-on tie, but I can screw up your journey for you; and they get to be responsible (at least in the narrow sense of that word) for spending a lot of money.

We need not, of course, apply this degree of thought to every highways alteration, whether by academics at a university or dim little highways officers. Speed cameras are rather different to other works in that they raise revenue and are not merely an expense. Nevertheless, it would be good to have some sort of risk/reward analysis made before money is laid out on what are often patently unnecessary works. As we head for recession, with billions in public money already pissed against the wall by low-grade local authority people, Oxfordshire’s highways officers are still tampering with road layouts with no perceptible benefit in mind, at least not one which could be described as necessary.

In that context, it is important to know that the government’s statistics on deaths prevented by speed cameras are seriously overstated. Just now, when money is short and when companies and individuals are having to make serious assessments as to what it is necessary to spend money on and what is optional, even, those high-spending but largely useless leeches who work for local authorities are going to start having to make serious judgements as to what really must be done. We don’t expect Exponential Bayes from them (the merest glimmer of intelligent thought would be a start), but elected councillors have some ammunition from the Liverpool study to begin questioning some of the highways officers’ assertions that this or that expenditure is really necessary.

The brighter policemen are beginning to get the point. Last week, Ian Johnston, President of the Police Superindents’ Association said that public confidence in the police had been “dented and bruised”, and singled out speed cameras as having erected a barrier between public and police. That has to be balanced against the recent bad news that Richard Brunstrom, the deeply stupid Chief Constable of North Wales known as the Mad Mullah of the Traffic Taliban, is not retiring this year as had been hoped. On balance, perhaps Mr Plod is beginning to learn that he needs some measure of public support and that street crime and burglary deserve at least as much of his attention as motorists going a little over the speed limit.

As for the Government, there have been occasional signs that even New Labour is wary of upsetting the large motoring lobby more than it has to and that criminalising decent people as well as soaking them in taxation is not a vote-winner. If Gordon Brown is looking for economies, he could start with the dead-heads in local authority highways offices. Perhaps he could start with Oxfordshire County Council’s highways department.

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: