Delivering the goods

On July 17 2007, the Public Administration Select Committee quizzed me for close to two hours on the content of my recent book Instruction to Deliver. The debate was lively and engaging. At a critical moment, David Heyes MP said sharply, ‘The most dismal result [of the recent Capability Reviews of government departments] is in delivery. Four years of [you] working on delivery, we do the reviews and delivery is a big failure area. Why is that?’

It is clear from the record that my first attempt to answer fell short: ‘difficult to understand,’ commented David Heyes quite fairly. ‘Your reason for the failure to deliver is that you moved the delivery target up too high. Surely that was a failure of what you were about as well?’

Under pressure, I finally found out what I really believed. ‘Exactly as you say, we raised the standard of what it takes to get successful delivery and then identified a gap. I do not say it was too high. The civil service is no different from any other organisation in an important aspect and that is that globalisation and technology are changing everything.

The British civil service is one of the best in the world but the challenges of the 21st century mean that the standards are being raised all the time. You are arguing that the expectations for the civil service were set by the delivery unit — by me — but they are set by the public. The expectations are out there. The frustration with delivery is out there. The public want higher standards and they want them now.’

This ultimately is why the government machine needed to become more effective at delivering, and why in the summer of 2001, at the start of his second term, the then-Prime Minister Tony Blair asked me to set up the prime minister’s delivery unit and issued me with his ‘instruction to deliver’.

My book describes my efforts to carry out that instruction between 2001 and 2005, telling the story of the crises we faced, analyses the politics of securing commitment to Tony Blair’s far-reaching reforms programme, and assesses the impact of those reforms. It also explains the demands of such a task from a personal point of view.

More important than the story, however riveting, are the lessons of that experience for the future of government not just in this country but across the world. After all, there is nothing uniquely British about facing up to the challenge of delivery. Every government in the world is faced with the daunting challenge of delivering ever-better results without demanding ever-higher taxes to pay for them. In other words, the productivity of public services is a central issue in politics the world over.

Moreover, citizens in the 21st century are very impatient, so politicians face a dilemma — they know they need a long-term strategy to secure significant change, but they also know that unless they deliver short-term results no-one will believe them.

In any democracy, if government fails to honour its promises or falls short of people’s rising expectations, then growing scepticism not just with a particular government but with the system in general is all too likely. So the stakes are high.

What are the lessons? First the delivery unit focused relentlessly on roughly 20 key priorities, including reducing health waiting times and cutting crime. In Blair’s first term there had been insufficient clarity and consistency about priorities. By establishing the delivery unit, Blair changed that. Moreover, we were not blown off course by the inevitable crises that beset any government.

Second, the delivery unit had a strong belief in the centrality of good, timely data. By insisting that those responsible for implementation of large programmes had good, up-to-date data (which, incredibly, many had not had in the past) we made a major contribution to evidence-informed policy-making.

For example, until 2001 the Home Office collected no data at all from police forces about recorded crime. From 2002 they began collecting monthly data. Similarly, the department of transport had no good data on road congestion in 2001 — but, as a result of our efforts, began by 2003 to collect excellent, real-time data on all ‘M’ and ‘A’ roads. The result was that these challenges could at last be managed and performance improved.

Thirdly, what made the delivery unit distinctive was that its task was not to answer the question ‘What?’, but ‘How?’ Answering the ‘What?’ questions has always been the stuff of government. Much of the government machine (and almost all media commentary) is devoted to answering them.

By contrast, the ‘How?’ questions (How will you get it done? How is it going? How do you know? How will you fix it if it is not working?) have been neglected.

The delivery unit was the only part of the government machine whose chief focus was to answer those ‘How?’ questions and we became expert at doing so. As a result, we were able to learn lessons from health that could be applied to education or criminal justice, for example. For the first time, knowledge transfer of this kind became systematic.

Beyond these practical lessons there were four underlying implications for government with general applicability.

Implication 1: Much of the time government is driven by crises and events but it is effective routines that deliver results. As Harold MacMillan famously said, the chief barrier to his success was ‘Events, dear boy, events.’ Through the vehicle of the prime minister’s delivery unit, we established routines for the prime minister which meant events did not distract from delivery. We ensured he received brief, crisp, monthly reports updating him on progress (or lack of it) within all his key priorities. We ensured he had quarterly ‘stocktake’ meetings with key cabinet ministers, where they discussed progress and solved emerging problems. And every six months we gave the prime minister and the chancellor an overall assessment of the likelihood of delivery of their priorities.

Without routines of this kind, governments fail. Blair himself believed this was the delivery unit’s single most important contribution. Between 2001 and 2005, Blair had many distractions from the delivery agenda that he could not ignore — September 11, 2001 and the Iraq war, to name just two examples — but he knew that, whatever he was doing, we were firmly focused on delivering his domestic priorities.

Implication 2: Reform of the civil service, so that it becomes capable of delivering for future governments of whatever persuasion, remains very much unfinished business. As he himself would no doubt agree, Tony Blair gave civil service reform insufficient attention in his first term. It was not until the middle of his second term that he began to give it priority.

He then urged departments to professionalise functions like finance and human resources, and encouraged the development of delivery skills. Much now depends on how Gus O’Donnell, cabinet secretary since 2005, builds on the welcome developments of his first two years as cabinet secretary, especially the capability reviews of government departments which have identified clear challenges in relation to planning and financial management, for example, as well as delivery. The test will be whether these problems are urgently and rigorously addressed in the next year and beyond.

Implication 3: Power is not a zero-sum — enhancing crucial relationships builds the capacity to get things done. By aligning key people behind a priority, by creating a culture where people are prepared to be open about problems before they become crises and by sharing the credit for successes, the extent of power — the capacity to get things done — is enhanced.

In other words, relationships matter. This is why, for example, it is important to build ‘a guiding coalition’ of leading politicians and officials who take charge of each major reform and share a deep understanding of the strategy. It is also why establishing effective working arrangements between the centre and departments (and indeed between departments and their key stakeholders) is vital to sustained success.

The delivery unit worked hard to create relationships with departments which were both challenging and productive. We understood that where a problem in a department was solved as a result of our work, it was important that the department got the credit. As US President Harry Truman explained, ‘It is amazing what you can accomplish if you do not care who gets the credit.’

Implication 4: The centre of government needs to become more effective and coherent. The demands on a modern prime minister are intense. The pace of decision-making is becoming ever more rapid, foreign policy increasingly demands the prime minister’s personal involvement and the 24-hour media is voracious.

The days when Churchill’s private secretary could find time to sit in the No 10 garden and read a novel or Attlee could fetch his own tea because he thought the messenger was ‘probably busy’, have long since gone. As Peter Hennessy has argued, from the late 1950s onwards the prime minister’s task has become ever more challenging.

In the demanding new century, a new, powerful and streamlined Department of Prime Minister and Cabinet (PMC) would strengthen the capacity of the prime minister to get things done on behalf of the government he or she leads.

The new department would simultaneously strengthen the civil service and the cabinet. In relation to public services I suggest within PMC, a permanent secretary responsible for performance (strategy and delivery) and a permanent secretary responsible for organisational health (people, civil service reform, machinery of government). Both would be answerable to the cabinet secretary. It could become more effective than the cabinet office has traditionally been by employing fewer, more effective, expert people.

Meanwhile, in the Treasury there should be a directorate responsible for public expenditure and productivity. There should also be a new cabinet committee focused on Public Services, Productivity and Expenditure (PSPE). It should be chaired by the prime minister, with the chancellor as an active deputy, key cabinet ministers as members, and jointly serviced by PMC and Treasury.

It would oversee a new two-year cycle for developing strategy, setting targets and allocating public expenditure. It would develop a new target framework, building on the developments of the last decade, and focus not just on hard outcomes but also on public satisfaction and commitment, efficiency and long-term capacity and capability. It would also review progress on delivery, as an equivalent committee does with real impact in South Australia.

I argue that the best constraints on prime ministerial power are the cabinet acting collectively, secretaries of state acting corporately and a civil service of such quality that the advice it offers is always taken seriously. These conditions have not been in place even in the recent past. Too often in my time in government, the civil service was seen as dragging its feet, a shock absorber rather than an accelerator. This meant that even when it was right to advise caution, its credibility was questioned.

Gordon Brown’s reforms at the centre take up a number of these suggestions as exemplified by the new permanent secretary role that Jeremy Heywood has in the cabinet office, the linking of productivity and public expenditure in the Treasury, the new PSA framework published in October 2007 and the new role for the prime minister’s delivery unit — to deliver all 30 of the new PSA targets — from within the Treasury. Of course, each prime minister should tailor the machinery of government to his or her priorities and style.

What is clear is that the role now requires a sharper, more central organisation with greater coherence, stronger relationships and more effectiveness than in the past.

Sir Michael Barber was the founding head of the Prime Minister’s Delivery Unit from 2001 to 2005. The author’s book Instruction To Deliver, published by Politico’s in June 2007, is available from www.westminsterbookshop.co.uk or call 020 7802 0018.