Civil service reforms never get killed-off, they just shrivel up from neglect until only a husk of the original is left.
So it has been with Next Steps agencies in UK government. What started out 30 years ago as a “revolution”, which at its height involved nearly eight out of 10 civil servants, has now dwindled to a barely recognisable administrative device involving fewer than one in four.
Next Steps was born in 1988 and was a programme of creating so-called executive agencies in Whitehall, named after a report entitled Improving Management in Government – the Next Steps. But its gestation went back a decade to Margaret Thatcher’s election as prime minister in 1979.
As part of her agenda for “rolling back the frontiers of the state” Thatcher immediately began an assault on what she saw as a bloated, inefficient, out-of-date Whitehall machine. Over the very first weekend of her new government, she appointed Derek Rayner from Marks & Spencer – where he had acquired a reputation for ruthless cost-cutting – as head of a new “Efficiency Unit” reporting to her as PM.
Rayner soon launched a series of efficiency scrutinies of Whitehall departments and functions. These were conducted by small teams of bright young things drawn from across the civil service who, over three months, would investigate some aspect of Whitehall’s function and make recommendations for savings. By the time Next Steps was published, more than 300 such scrutinies had been conducted.
Three years after the Efficiency Unit was established, the government launched another major drive to cut costs and improve efficiency: the Financial Management Initiative. This was an attempt at a more systemic change – to make civil service managers at all levels responsible for delivering their objectives through the most efficient use of their resources, especially their budgets.
The problem was that neither of these schemes delivered results in their first years. In March 1986, the National Audit Office found that of £215m of scrutiny “savings” identified, only £51m had actually been delivered – less than a quarter. FMI was also widely seen as being bogged down, unable to provide managers with the information or flexibilities to deliver the efficiencies that were being demanded from them. Another NAO report in October 1986 was critical of FMI, but mainly concluded it was “too early to tell” if it would work.
It was then decided upon that there should be a sort of scrutiny of scrutinies, and FMI, by the Efficiency Unit. Kate Jenkins, Karen Caines and Andrew Jackson from the unit conducted the study from November 1986 through to March 1987. The result was the Next Steps report which was duly submitted to the prime minister.
The report has often been wrongly attributed to Sir Robin Ibbs, the then head of the Efficiency Unit. But the names of Jenkins, Caines and Jackson appear on its cover.
Its central conclusions about the existing attempts to reform the civil service were clear: they were not working. It listed five areas that were “serious problems” and concluded something more radical and systemic needed to be done.
While the prime minister apparently accepted the main thrust of the report, given to her in early 1987, it was deemed too sensitive for immediate publication. Thatcher was gearing up for a general election in June 1987 and broadcasting a report saying her civil service reforms were not really working was not seen as a good idea. Publication was delayed until early 1988.
What was the Next Steps policy?
The central recommendation of Next Steps was an organisational upheaval in Whitehall: “The aim should be to establish a quite different way of conducting the business of government.”
The report said: “We recommend that ‘agencies’ should be established to carry out the executive functions of government within a policy and resources framework set by a department.” The “central civil service should” it added “consist of a relatively small core engaged in the function of servicing ministers and managing departments”. Responding to these departments would be “a range of agencies employing their own staff, who may or may not have the status of crown servants”.
The purpose was to give “the management of [each] agency…as much independence as possible in deciding how [its] objectives are met”. That should include “freedom to recruit, pay, grade and structure in the most effective way” within their policy and resources framework.
The idea of establishing executive agencies to deliver services was not entirely new. The Fulton Commission in 1968 recommended that the government should conduct an “early and thorough investigation” of “hiving off” many civil service functions along similar lines to that long practiced in Sweden, which members of the commission had visited and were greatly impressed by.
As an aside, Fulton’s recommendation was never acted upon in Britain. But in 1971 the civil service in Hong Kong, which was then under British control, was divided between policy branches and executive agencies along very similar lines to Sweden. There is no direct evidence about whether this was an experiment, but it does seem like a rather big coincidence.
The Next Steps report took a broad, pluralistic, view of what might constitute an agency and whether specific agencies should be inside or outside government. It also thought legislation might be needed to establish an agency’s proper purpose and accountability framework. Further, the authors thought that the agency structure “could be used to cover a substantial proportion of the activities of the civil service”.
“The idea of establishing executive agencies to deliver services was not entirely new: 1968’s Fulton Commission recommended hiving off many civil service functions along Swedish lines”
Not everything that was recommended in the report made it into the implemented Next Steps programme and some other changes appeared much later as it unfolded, rose and then fell into neglect.
What happened next
When the Next Steps report was submitted to Thatcher in March 1987 it set a very bold timetable. It proposed that “within two years at the most, departments should have completed identification of areas where agencies are the most effective way of managing and should have changed their own internal structures to implement this change”.
Although the report had not been made public, or the policy announced, Whitehall set about the process of implementing this radical restructuring over the summer of 1987. By October, 12 candidate agencies had been identified, of which the largest by far was the Employment Service with 35,600 staff. The smallest was the QEII Conference Centre, with just 65 staff.
In February 1988 Thatcher finally announced the programme to parliament. A full (second) permanent secretary role had been created in the Cabinet Office to run Next Steps and Sir Peter Kemp, a former Treasury official, was appointed.
The rise of Next Steps
The first three agencies launched in 1988 were the Vehicle Inspectorate, Companies House and HM Stationery Office – hardly Whitehall giants. Between them they had only around 6,000 staff – 1% of the civil service.
It took about a decade, not two years as originally envisaged, for the Next Steps agency-creation programme to complete. The number of agencies steadily increased until a peak of about 140 agencies (including organisations like Inland Revenue, who were said to be operating “on Next Steps lines”) in 1998. Many of these were very small and for most of this period the “big five” – Inland Revenue, Customs, Prisons, Benefits and Employment – employed around two-thirds of all agency staff. By the time New Labour came to power in 1997, three out of four civil servants worked in agencies.
There was resistance to the roll-out of agencies. In 1992 – only four years in – Kemp, the driving force for implementing Next Steps, suddenly left the civil service. When he published a pamphlet on Next Steps the following year it became apparent he had not gone willingly. He wrote that “at the moment the levers of power [in the civil service] are in the hands of those who broadly oppose reform”. He went on to say that “the guards are in charge of the escape committee.”
Kemp’s critique was clearly exaggerated, as the numbers testify. But in some respects he was right. The original radical version of Next Steps got lost in translation.
“The original radical version of Next Steps got lost in translation. As the programme developed it rapidly became a one-size-fits-all enterprise”
Next Steps had originally proposed a pluralistic approach to hiving off agencies, creating a variety of different arm’s-length bodies, but as the actual Next Steps programme developed it rapidly became, formally at least, a one-size-fits-all enterprise.
Every agency remained part of the civil service, as opposed to some being placed outside, as the report had suggested. None were established by legislation – they were all merely administrative units within government departments that could be created, or abolished, at the whim of ministers. Most agencies had a chief executive, who supposedly reported directly to ministers, although many smaller agencies in practice did not; a framework document and performance targets formally agreed by ministers, although often not in practice; a resources agreement; and some freedoms about pay and grading.
Only in one important respect did Next Steps go further than the original report in practice: agency CEOs became an accounting officer for their agency. This meant they could be summoned to parliament to answer questions about their agency and be formally held to account. Their permanent secretary was also an accounting officer, including for their agency, which meant each agency had two accounting officers.
One unexpected consequence of this was that some agency CEOs, especially of the largest agencies, felt able to speak out fairly frankly when giving evidence to parliament and that many more civil servants – sometimes at much more junior levels, such as the CEOs of small agencies – were summoned to parliament.
The fall of Next Steps
One major innovation within Next Steps was the amount of information that was disclosed about the inner workings of Whitehall. Every agency produced its own annual report - although they weren’t always easy to get hold of for some of the micro-agencies - and a plethora of other published material.
In addition, the Next Steps team in the Cabinet Office collated and analysed a lot of this data in a compendium Next Steps Review, starting in 1990.
A picture of the 10 published annual reviews tells an interesting story. As the number of agencies grew, so did the compendium annual report. And the analysis they contained got more and more sophisticated. Details of trends in agency performance over time began to emerge – just the sort of performance and accountability that was supposed to be a central part of Next Steps. The data and analyses were a cornucopia for those of us studying government. It didn’t last.
In 1999 the format of the review changed dramatically and it became a fraction of the size of its immediate predecessors. Many smaller agencies were excluded from the review and the analyses were much more superficial. Despite the performance of agencies supposedly being central to New Labour’s “Modernising Government” agenda - according to the ministerial introduction - this was the last time the agencies’ annual review appeared. It was replaced with a review of some agencies and quangos, but with little of the former detail.
Over the next decade the agency organisational format gradually faded. Many of the supposed freedoms were slowly reduced. Some of the delegations over things like pay and grading were pulled back into the centre. New Labour’s focus was on “joined up government”, not unbundled government.
The number of agencies gradually declined from its peak of around 140 to around only 40 by 2013 and has remained near this level since then (the figure was 37 in 2018). Some of the reduction in organisational numbers was because of a wave of agency mergers – especially after the Coalition government took office in 2010, having pledged a “bonfire of the quangos”.
But staff working in agencies also reduced dramatically as many former agencies were also re-absorbed into normal departmental structures. By 2018 only 25% of civil servants worked in agencies, down from 78% at its highest.
Did agencies work? Did they improve management in government? At the risk of sounding too academic, the answer is probably yes and no. The programme was never as revolutionary as originally envisaged and became over-bureaucratised. It certainly did shake up some things in Whitehall and led to improvements in efficiency, effectiveness and customer service in many cases.
But like many civil service reforms it relied on the support of ministers to really make it work. And as all too often happens, ministers change or lose interest. Perhaps if it had been more institutionalised – embedded in legislation for example – Next Steps may have been more successful.
As it is, it’s just one more reform that left traces of its once high hopes, but not really much else.