Early in 2002, Tony Blair addressed a small group of officials working on the top 17 delivery priorities for his second term as prime minister. “You’ve probably got used to the idea that your career depends on me,” he told them. “I’m just coming to terms with the idea that my career depends on you.”
Less than a year since a general election that he interpreted as an “instruction to deliver” public service reform, Blair was acknowledging the delicate relationship between politicians who must set the goals, and the officials whose job it is to reach them. His remarks, according to historian Dr Michelle Clement, were an attempt to “embolden” the civil service, through the relatively new team he was speaking to – the Prime Minister’s Delivery Unit.
Headed by Michael Barber, a former special adviser in the education department who became a civil servant and then moved to No.10 to create PMDU, the team’s role was to track and support progress in four broad areas: health, education, home affairs and transport. Clement – a lecturer in the Strand Group at King’s College London and researcher in residence at No.10 – has spent more than a decade studying the way this unit operated. Her study has been informed by 600,000 words of diary entries written by Barber between 2001 and 2005 (see box), alongside interviews with other key officials and politicians – required, she says, to offset the fact that a diary inevitably centres its author.
This work is the basis of her first book – The Art of Delivery: The Inside Story of How the Blair Government Transformed Britain’s Public Services – which explores the “human machinery of government” at the heart of New Labour’s reform agenda. It offers new perspectives on Blair’s second term, and it feels very timely. Like New Labour in 2001, the current government has a political imperative to deliver public service reform. And like Blair, prime minister Keir Starmer has turned to Barber for support – appointing him as an unpaid adviser on effective delivery. A Mission Delivery Unit has been established in No.10, headed by Clara Swinson, a former member of PMDU. Meanwhile, in the Department of Health and Social Care, a Secretary of State’s Delivery Unit is being set up.
Health secretary Wes Streeting has also appointed his one-time predecessor Alan Milburn, who worked closely with Barber, as the department’s lead non-executive. Milburn features heavily in Clement’s book – through her interviews, we see him move from being somewhat sceptical of the new PMDU – worried in part about confused accounting lines between it and the Treasury – to describing it as “fantastic… a gift”. We also gather plenty of insight into his thinking on health reform and his approach to change as a politician – insights which are interesting in themselves, but have particular pertinence given his current role at the department.
Most civil servants are probably familiar with Barber’s approach. His science of delivery – named “deliverology” by Treasury grandee Nicholas Macpherson – has been adopted by governments around the world. Broadly, it involves the careful setting of targets linked to real-world outcomes, and the development of plans based on data and the partners required to make change. These are monitored by regular stocktakes which track progress, spot problems, and bring the power of the prime minister to bear in solving problems and pushing for more change.
Clement makes a distinction between this science and the “art” which underpins it. When she began working on the project it was “peak science of delivery”, she recalls, and there was a sense of “just follow this list, and you’ll get the desired result”. But she had questions. How did Barber navigate the Blair/Brown dynamic which coloured so much of the New Labour period? How involved was Blair, really? “There was this idea that as Blair focused on foreign policy post-9/11 and then with the Iraq War, he became less focused on domestic policy,” Clement says. Yet since the domestic reform agenda did carry on, with strong success in key areas, she adds, “that idea didn’t quite feel right to me”.
As she analysed Barber’s diaries and spoke to those who worked with him, she sought to draw out the “human dimension to [deliverology] – the messy uncertainties”.
“It leaves me with the problem (which I relish) of winning the approval of two irreconcilable positions”
Michael Barber, writing in his diary
The science of delivery is ineffective without this human dimension, she argues, describing the art of delivery as the “shared history, deep understanding of political context and ability to build cooperative relationships which oils the machinery of government”. The approach of using diaries and interviews of course gives a different insight than studying official announcements or records for this period (many of which are not yet released, so Clement has not used them for her work). Reflecting on a meeting about asylum policy in 2001, for example, Barber writes that from the official minute you would think it was an “ordered, rational event”, but “in fact what happened was an unholy row”.
It also gets under the skin of how the Delivery Unit really made an impact. Alongside the science, there was a strong focus on building relationships and navigating the conflict inherent in many discussions on the interface of politics and policy. The whole PMDU team undertook conflict resolution training, for example, and would invest a great deal of time preparing not just data for stocktake meetings but also understanding the incentives and positions of all those present. Swinson reflects that building trust was “completely fundamental” to their approach, as was the need to create a guiding coalition of partners who would deliver change.
Barber’s personality played a key role in all this. Swinson describes him as “at heart an educationalist” whose first instinct was to learn from any challenge, while Clement tells CSW that “he just didn’t rise to aggression”. Clement explains, “It’s not that he didn’t feel it internally and sometimes reflect on it in his diary, but he had a discipline of mind that would make him focus on how do we get through to a solution.” This capacity to reflect on, rather than rise to, conflict was apparent when a critical PMDU memo on the health service was leaked in 2003. Barber wrote at the end of the week that Milburn had been “unhappy but restrained” when discussing the leak. Milburn, speaking to Clement two decades later, says: “I think that’s a very, very kind interpretation of that conversation... I remember swearing at him, along the lines of, ‘What the fuck is this? Why have you created this problem?’”
How Barber's diaries became history
Barber wrote his diaries, Clement says, from “the perspective of a historian by training”. They formed a valuable way to reflect on events and develop his own thinking, but he was also conscious they could provide deep insight for those who study government.
In 2011, he approached Jon Davis – now a professor of government and director of the Strand Group at King’s College London – about the possibility of using his diaries as the basis for a PhD. Clement, who had just completed her MA under Davis, was chosen to work on what she describes as an “extraordinary primary source”.
Having completed her PhD in 2020, Clement now teaches on the MA in Government Studies at KCL, and is also researcher in residence at No.10. “The main aim of the latter position is demystifying the role of prime minister and the office of No.10 for the public,” she says. This includes writing blogs and giving talks and hosting KCL classes at No.10, but she also brings the expertise she has gathered through her academic research to, “wherever possible, help those behind the scenes”.
A unifying theme in all of Clement’s work is the view that “institutional memory in government is important and lacking”, and that understanding how previous administrations approached the challenges of government can bring real benefit to those still facing those challenges today.The government has pledged to build 1.5 million new homes by the end of the parliament – or 370,000 homes a year, well above the number built in recent years. It has set mandatory targets for local authorities in a reformed National Planning Policy Framework, which also requires councils to review green-belt boundaries and identify lower-quality “greybelt” land that could be made available for new developments. If councils are failing to meet their targets, central government will step in to oversee delivery.
Barber’s ability to navigate conflict and build consensus was vital for coping with the increasing tension between Blair and Brown. Clement details how Barber had to manage this right from the start, fending off the risk of the new unit being squashed by a Treasury suspicious that No.10 was trying to undermine the chancellor’s public service reform agenda. While Jeremy Heywood and Ed Balls played the role of representatives for their principals, brokering deals together before presenting them to Blair and Brown, Barber was trying to build common ground between very different perspectives. In 2003, during a debate over the creation of foundation hospital trusts – Blair was a supporter of the idea but Brown was wary – Barber wrote in his diary: “It leaves me with the problem (which I relish) of reporting in a way which wins the approval of two irreconcilable positions.”
Barber developed a good relationship with Brown, bonding over their shared training as historians. Clement says Barber could “see the intellectual and the institutional power of the Treasury and how much progress they had made in the first term. He realised they had to find a way to bring PMDU together with a burgeoning Treasury with a big social fairness agenda”.
By giving us a peek at how deliverology unfolded in real time, Clement offers not just historical insights but also practical gems for civil servants ready to learn from the past. We observe how Barber secured sufficient prime-ministerial time as foreign policy loomed large first in 2001 and then in 2003. We learn how he solves both policy and relational problems with patience and the occasional bravura move. And we follow the twists and turns of negotiations around the delivery contracts, spending reviews and policy announcements which PMDU supported.
Are there any broad lessons, CSW asks Clement, for officials currently thinking about how to drive delivery? Clement believes the current Labour administration has “come in with a bolder agenda to begin with” than the Blairite government. But she adds a challenge which emerges from studying a significant period of change in government. “You have to accept that what will be delivered in the next four years will be largely delivered by the system as is. You will be able to make changes that will impact your second term, or even a third term. But you’ve got to be very clear minded about the levers that are available to you now, that can be created rapidly, and about how you motivate people to make change quickly.”
To do this, she suggests, officials should focus on the human elements and build solutions that are mutually beneficial. “Get the right people in the room: the ministers involved, the senior civil servants involved, representatives from the Treasury; do the preparation, agree the data beforehand, work out where the blocks are… and deal with it, using prime-ministerial power to hold civil servants and ministers to account.” Clement says she doesn’t think there’s a recognition that these things were happening under Blair. “There’s so much focus on sofa government or bilaterals – and, of course, that did exist. But there was also a real effort to make government efficient and get the right people in the room.”
There’s a moment at the start of Blair’s third term where Barber observes his approach adapted to something more like a cabinet committee, with Brown present. Rather than focusing on the delivery priorities at hand, Blair and Brown argued over philosophy. Reflecting in his diary later, Barber wrote: “Blair doesn’t want to have to defend himself from Brown and Prescott… he wants to drive his programme through.”
“You have to accept that what will be delivered in the next four years will be largely delivered by the system as is, and be very clear minded about the levers that are available to you now and can be created rapidly”
Michelle Clement
In 2001, Barber told Heywood – then principal private secretary to Blair – that his “‘ambition was to work the Delivery Unit out of a job” by achieving the priorities they had mapped out. In the end, the unit has had a much longer legacy than this. “It was originally created as a machine for the prime minister, specifically for the length of a parliament,” Clement says. “But it became mutually beneficial for many people in the system. As time went on, they realised: this is useful for ministers, this is useful for other departments that aren’t directly involved; it’s even useful for the chancellor and Treasury.”
This shift is evident in a comment made by Home Office official Martin Narey in 2003. During a meeting on drugs strategy, the commissioner for correctional services said: “The disciplines the Delivery Unit had advocated were beginning to work, and it now seemed incredible that we managed for so long without them.”
Over its four years, PMDU kicked off a shift in the way Whitehall thought about delivery. Since the late 1980s, delivery capability had largely been focused on “Next Steps Agencies” – named after a report created for Margaret Thatcher which recommended moving responsibility for operational matters into arm’s-length bodies rather than delivering through departments. At the start of the period, Barber characterises civil servants in the transport department as “intelligent, urbane officials who believed they were in no position to deliver anything”. By bringing back a culture of delivery around key priorities, Clement argues that Blair and Barber built an “enhanced capacity to deliver within Whitehall”.
Barber left PMDU in 2005 and it lost some influence, until it was eventually disbanded under David Cameron. Since then, other delivery and implementation units have emerged with different structures. “It is an odd situation that prime ministers find themselves in, without a department,” says Clement. “Their home secretary or foreign secretary or chancellor has thousands of civil servants, whereas the centre can often be underpowered. But the expectations on the centre are huge. So it’s interesting to see how different prime ministers have adjusted and adapted to that situation.”
Ultimately, she says, most come to realise they do need something to track and push through priorities. “A delivery unit can be – as it was [under Blair] – an extension of the prime minister, or you can have [one] that is more like a service – something like the Government Digital Service.
“It’s not the same beast,” she adds, “but I think that’s how we’ve seen delivery units evolve in the UK – they became de-politicised. And really, to be most effective and sharp, you can’t get away from the politics that is involved in governing. That’s what PMDU managed, quite skilfully.”
In 2005, Blair made another visit to PMDU, just before Barber was about to step down as its leader. Barber, reflecting on the role the staff played in driving Whitehall towards the PM’s priorities, said to him: “These people are your street fighters.”
“Yes,” Blair replied, “but they do it with such charm.” The PMDU staff, Barber records in his diary, were “delighted”.
The Art of Delivery is available to pre-order now