Last month saw the publication of a report by the Social Market Foundation (SMF) entitled “To the point: a blueprint for good targets”. The Minister for the Cabinet Office, John Hutton MP, launched the report. This special issue is devoted to what the Minister and report had to say.

Normal service will be resumed next month.


“What is a good target?” Was the title of the Minister’s speech. He started by saying how pleased he was that the SMF have recognised that targets do indeed have an essential role to play in delivering better public services. He would be pleased; he is a politician. If the report had said targets were making the public sector worse he wouldn’t have been there to speak.

The Minister said: “My view is that targets, properly constructed and applied across public services have been fundamental to past successes and will be an essential part of sustaining progress into the future.” So it begs the question: how do you ‘properly construct’ a target? Of course one cannot find the answer in the Minister’s speech. Like many, he behaves as though questioning targets simply cannot be countenanced.

To those who think that targets should be abandoned he said:

“There are those who want to scrap all of the targets that have been set by Government for the public services. This betrays a total lack of confidence in public services and their ability to offer more responsive, personalised services.”

How does scrapping targets show a lack of confidence in public services? On the contrary, I know a number of public services that have been improved by, amongst other things, scrapping targets; the leaders learned that targets were part of the problem and not part of the solution. To offer a responsive personalised service you need to learn to design against demand, targets stop you doing that; it is in their nature.

And he said:

“It would be entirely the wrong thing now – as some people have argued – to abandon every single target that has been set for the public services. The real issue facing us is not whether to have targets. The real issue is what are the right ones.”

But he didn’t give an answer to the question, what are the right ones?

Instead the minister gave a series of justifications for targets:

“All governments over the past 25 years have introduced elements of national target setting.” Does that make it the right thing to do?

“Our performance management regime as it stands is mainly focused on providing accountability from local delivery bodies to the centre. Some of that is understandable – people expect the ministers they have elected to be able to effect change in their local services and hold them responsible if they do not.”

But the means by which ministers attempt to effect change actually makes services worse; when the truth is out will he stand up to be held responsible?

“ People are entitled to have a means to judge whether services are improving or not and whether providers of these services are responding properly to the needs of their customers and consumers. Targets help do all of these things.”

No they don’t. Targets are not a reliable method for assessing a service from either the users’ or manager’s point of view. People make judgements about public services from the experiences they have with them, and these experiences are currently governed by the need to meet targets. There is a host of examples where the service looks good to the target-setter but rotten to the customer.

“In every other part of the economy we see performance targets accepted as a mainstream device for managing organisations and complex systems.”

I think it was Socrates who said you won’t find the truth by counting heads. And in plenty of parts of the economy, both public and private sector, managers are learning that using measures derived from the work leads to improvement you would never have conceived as an achievable target. And, by the way, targets make systems more and unproductively complex, something he does not know.

“We do need to recognise their limitations and not be blind to the perverse incentives and gaming that they can promote within an organisation. But it is precisely because of their power and leverage to effect change that we need to be sensitive and increasingly skilled in their deployment.”

So its back to ‘its OK if you do them right’, but no advice on how to do that; instead an expectation that the perversities will have to be managed in some way.

The Minister claims targets have improved waiting times, adult learning and crime. But I would bet him my bottom dollar I could prove otherwise. In every public sector service we have worked we have found targets to be a principal cause of sub-optimisation; yes they make the work worse. And to re-design for improvement meant taking the dysfunctional consequences of the targets away. I bet his conclusion is based on the data reported upwards and as we know with all of the services we have worked in, these data are unreliable and invalid. The Minster sits atop a management factory of vast proportions, ignorant of the reality of what is happening on the ground.

Despite being unclear on how to set targets and yet feeling justified that their role is essential, the Minister went on to say that changing the way that Government sets targets has already started. This can only be recognition that there are problems, but we only get clues as to what these problems are.

He said there is a need to avoid unintended distortions to service delivery. But we are not told how ‘better’ targets will do that. He said there are already fewer targets – a fraction of the original number. So if there were problems with a lot of them why would we not have the same problems with fewer?

And he said we need to consult: “It is also right that we continue to consult with front line workers and the public to ensure that targets are reasonable and achievable, that measurement regimes are proportionate and that the targets take full account of the other reforms that are under way.”

If any of those who have improved their services to levels beyond what was thought achievable as a target thought this way they never would have improved. Instead they learned to design against demand. People want services that work from their point of view, designing this way delivers better services at lower costs, something the Minister wants but only achieved by ignoring his advice and guidance.

The scariest part of his speech was his view of the future. He wants to hand power to individual service users. People will be able to choose between public service providers; he thinks the market will drive improvement. We are, for example, seeing the introduction of the ‘choice’ to make your own payments for adult social care, old people being given cash to pay service providers directly. Aside from the complexity of the tax implications this puts one of our most vulnerable groups into a position they really don’t need or want to be in. Right now we find adults requiring social care to be subject to massively long waits for assessment and provision – it can take so long to get the care that people get worse. Instead of remaining independent, people become more dependent – exactly the reverse of the Ministers’ intentions.

It might not surprise you to know a central part of the problem in adult social care is the targets and specifications regime that sits over the work. The scope for improvement is considerable. Making old people make ‘choices’ won’t solve the problems of the current design; people want services that work, making them work better also makes them cheaper. To make them work you have to remove the arbitrary measures and instead use measures derived from the work. Something the Minister ought to take an active interest in.

Instead the Minister tells us we can look forward to something plausibly masquerading as ‘best practice’. He announced that he is going to develop a way to measure customer satisfaction across our public services. “If we can develop a workable customer satisfaction index, it could be hugely useful in ensuring the targets of the future are closely aligned to the priorities of service users.”

If only he knew.

The Minister finished by saying he was “very open to engaging in the sort of debate that the SMF work will stimulate.”

I hope he is true to his word. I am going to send him a copy of this newsletter.

Now I turn to the Social Market Foundation’s report.

The report promises “an objective, cool, fact-based, look at the arguments”. So let me summarise the arguments discussed and the response from the report ’s authors.

Argument: There are too many targets.

Response: A lot is not automatically too many. Reduce the number. Rank order their importance. Use them principally when something badly needs addressing.

This is the less is better argument. Less of a bad thing is still a bad thing. And the striking and illogical conclusion – to only use them when something badly needs addressing does not follow. It becomes, however, the main recommendation.

Argument: Targets are too rigid and undermine the morale of staff.

Response: People will not always like being held to account; there is no need to lament conflicts that might arise. If it can be shown the target is sensible and rational, then this is fine. The ultimate authority rests with government hence there is no problem with targets being centrally set.

In other words, tough!

Argument: Targets have perverse, unintended consequences.

Response: Think hard about how to devise targets that minimise this behaviour; perhaps the best strategy is to expect it. Devise targets with the following malign assumption in mind; that everything legal, and a few things not, will be tried to keep up the appearance of having hit the target. Have greater transparency about the reasons for selecting targets. There is no reason why they should lead to distorted priorities. This won’t happen if we have the right target. The very fact that professionals are asked to do one thing rather than another is deliberate, it may be a problem but it is intended. If a target has focused on the wrong priority, this is a case for changing the target, not abandoning it.

I guess that means act without theory, do your best to manage the problems and expect it to be tricky.

Argument: It is not always clear who is responsible for a target.

Response: The very point of a target is to determine a desirable outcome and assign responsibility for it. Where targets involve more than one agency they need to be sub-divided into contributory targets. It must always be clear who is the lead department.

So make someone responsible.

Argument: The data are not always reliable.

Response: Reform the agencies producing and validating the relevant statistics. (A core component of the blueprint for reform). The problems with data are measurability, quality and vulnerability to manipulation. Establish standards for measures. Determine degree of external audit.

This means more costs in specifications and auditing.

And so to the report’s recommendations

Targets must be set only when change is required or for aspects of public services which are exceptionally important – not when the status quo needs to be maintained or standards upheld. Thus, targets are best used in times of service failure, where there is a policy decision to significantly raise performance standards, or where there is a consensus that a service needs to be delivered in a radically different way: they must identify what is not working in a service, and what is of national importance.

Consequently, there should only be a handful of targets in each service area, which are dynamic depending on service performance at any given time.

Eh? Did I miss the logic? What do we do with all the others?

The report has no logic. In its introduction the authors promised to scrutinise “the standard – and some non-standard – criticisms of targets, distinguishing the valid ones from those that appear to be driven at least in part by professional self-interest.”

I didn’t see any of my arguments scrutinised. They are (as you know from newsletters passim ad infinitum):

There is no reliable method for setting a target Targets ALWAYS sub-optimise systems using measures derived from the work ALWAYS leads to levels of improvement that would never have been conceived as achievable targets

I am ready and at the minister’s service to explain all three.

A good target is an oxymoron. Targets do not pass the test of a good measure: can you use this to understand and improve performance? System measures pass this test and the evidence of their value in use is incontrovertible. The minister should be told.