W ITH AN election looming, public services and their improvement - or lack of same - have climbed to No 2 on the political agenda just below Iraq. As with foreign policy, it promises to be a white-knuckle ride. For while there's clear water between the two main parties on the public sector - the government's focus is on improvement, whereas the Tories' is on cost-cutting - ministers face a growing problem of evidence.
Basically, voters refuse to believe public services are improving as fast as the government says they are. Unease at the 'perception gap' is almost palpable. Nick Raynsford, the local government minister, has already puzzled over the fact that local authorities' customer satisfaction scores are going down at the same time as their official performance indicators are going up. The same theme was a subtext to many of the presentations by Whitehall bigwigs at last week's Economist conference on the public sector.
For every area of policy, the conference heard, almost all the government's traffic-light performance indicators are green. There are more doctors, nurses, policemen and teachers. Crime is down, school achievements up, NHS waiting lists are shorter. Speaker after speaker told the conference that from where they were sitting, services were getting better - it was only the extent of the improvement that was in doubt.
What's more, as the presentations documented, this is in the perspective of a government for which public services really are important. They are seen, rightly, as an essential component of a competitive economy and, at a guess, No 10's performance management system, at least in form, is by some distance the most sophisticated ever put in place.
So why isn't it delivering? Identifying answers has become a Whitehall industry in itself. One suggestion is poor expectation management: people expect too much. Another is a time-lag between the personal ('the hospital treated me quite well') and the general ('the NHS is getting better'). For Raynsford, more bizarrely, it appears to be something to do with a lack of public leadership.
But there is a much simpler explana tion to hand. The signals ministers are receiving do not mean what they think because they are not transmitted by service users but by service managers. They reflect a corporate rather than a public experience, not the same thing at all. Yes, it's the old problem of targets.
Last summer a borough chief executive predicted to me the divergence between his government and public satisfaction figures. 'The targets and specifications handed down from the centre oblige us to do things the public don't care about,' he explained. A striking example is e-government - web-enabling public services. In the latest round of specifications, councils are encouraged to make it possible for citizens to browse their council tax online. 'How sad would I have to be to want to browse my council tax payments online?' mused another service director.
Across the field, public service managers are diligently ticking off and reporting what matters to government, not to citizens. This explains baffling (to the public) rankings like three-star NHS trusts or 'excellent' councils. 'Improvement' or 'excellence' is in the eye of the government, not the public. Hence the gap.
But this is in the very nature of targets. The particularly ingenious Catch 22 to which they are subject is so well documented it has a name. The social-science equivalent of the uncertainty principle in physics, Goodhart's Law states that the instant a measure is used as a target, it loses all value as a measure.
This is because managers understandably devote their efforts to meeting the target, not what the target stands for. Targets, as Michael Barber, director of the Prime Minister's Delivery Unit, helpfully reminded the conference, are 'representations', abstractions from the aim beneath. Indeed: and that's the problem.
It's sensible that all A&E casualties should be treated as quickly as possible consistent with clinical need. But as soon as that is represented as 'all emergencies must be seen within four hours', as currently mandated, doctors divert attention from juggling patients according to clinical need - which may mean operating on one person within two minutes and leaving another for six hours under observation - to cramming everyone through the arbitrary threshold to avoid a waiting-time 'breach'.
Where the target and common sense are in conflict, staff employ a variety of recording ruses to reconcile them. As Goodhart predicts, the measures are no longer reliable - doubly so, since the target was an inadequate representation of a complex aim in the first place.
Compounding the problem, as Lucy de Groot, executive director of the Improvement and Development Agency, reminded listeners, is that 'the target regime is derived from silos' - individual departments or agencies, or even sometimes departments within departments. Being un-joined-up, they fail to coincide with the lived experience of citizens.
At worst, as documented in a previous column, that leads to a situation where a water company has measures in place for how long it takes to answer the phone or make an appointment, but none for the end-to-end time to fix a burst main. It can meet all its service standards - and thus proudly claim 'excellence' - while it leaves the water running for weeks.
The government is now well aware of some of the shortcomings of target regime. It has cut their number from about 700 to 100 departmental Public Service Agreement targets, according to Barber. A full-scale assault is being launched on the pounds 11 billion regulation industry - an initiative welcomed by Audit Commissioner James Strachan, a champion of the need for regulatory value for money.
And it is saying all the right things about getting away from top-down control and giving service deliverers the freedom to deliver 'personalised' services. Former Treasury adviser and prospective MP Ed Balls suggested that the Bank of England should be the model for future public sector reform, offering a stable framework within which specialists could make decisions unencumbered by short-term politics. 'One of the things we've learnt is that we need to get systems, rather than individuals, right,' he said.
But the government still cannot help getting targets in a twist. Witness the new goal of halving MRSA infections in hospitals in the next three years. Think what this looks like from the patient's point of view: in three years' time, I will have half the chance of being killed by going to hospital than I do now. The only acceptable 'target' here, as in all such cases, is perfection, and the only acceptable measures those that show both public and providers how progress towards it is being achieved, year by year.
The Observer, 14 Noember 2004