+44 (0)20 7799 6677

How many civil servants should it take to run the country?

Tim Ambler, 1 December 2022

Last May the government, with Rishi Sunak as Chancellor, decided the civil service should be cut back to the size it was pre-Brexit.[1]  That would be a reduction of about 91,000 or £6.9 billion growing annually with salary increases not to mention (as government rarely does) the whopping unfunded consequential pensions. That remained the policy until the November Autumn Statement when, although we were told the spending bucket was being scraped for savings, that one wasn’t. In fact, the Prime Minister had already reneged on his undertaking. He announced on 1st November that he did “not believe that top-down targets for Civil Service headcount reductions are the right way to do that. ‘Instead, the Chancellor and I will be asking every Government department to look for the most effective ways to secure value and maximise efficiency within budgets so that we can use taxpayers’ money sustainably in the long term.’” [2]

It’s early days and he may not yet have got the hang of being Prime Minister but setting targets for departmental heads is how things get done.  In fact, it seems that the Treasury civil servants had persuaded the Chancellor that the redundancy costs would wipe out the savings. As the April plan was to make the savings over time mostly from natural turnover, the advice was misleading. Anyway, leaving it all to departments means that reducing civil servant headcount is the very last suggestion Permanent Secretaries will make, if at all.

The real issue is not staff numbers but departmental efficiency.  Back in 1988, the then PM, Margaret Thatcher, announced that government departments should consist of only two types of body: the central core and executive agencies which would focus on delivering the department’s missions. Executive agencies would have measurable goals, headcounts and other costs and report annually to their departments. Mrs Thatcher, and Robin Ibbs, her chief adviser, had decided that the big thinkers who decided policy and proposed legislation were different sorts of people from the effective managers needed to deliver them.[3]

The civil service hated this “Next Steps” reform and devised all kinds of ways to frustrate it. The Ibbs team focused on getting the policy agreed, which was exceedingly difficult, rather than making it happen. Cabinet Secretary Sir Robin Butler, as he was then, commissioned a review in 1991 of how Next Steps gained acceptance and provided the Foreword but Diana Goldsworthy’s admirable account[4] is more about the politics of gaining acceptance than the rightness of the concept itself.

As at 31st March 2020, government was admitting to 295 “Arm’s Length Bodies”: 38 executive agencies, 237 non-departmental public bodies (NDPBs or quangos) and 20 non-ministerial departments (NMDs).[5]  There are many more, public corporations, for example, like the Bank of England or the BBC, but the two categories that should have been killed off by the Thatcher/Ibbs reform are a key part of the failure to change.

NDPBs and NMDs have one thing in common: they are part of, but also independent of their host departments.  They nominally report direct to parliament (but no one knows to whom) and in practice answer to the Secretary of State of the host department who appoints their members. The NHS is an executive NDPB. HMRC is an NMD and reports “to Parliament through our Treasury minister”,[6] i.e. one of the ministers of HM Treasury.  NDPBs are a cunning civil servant ruse for getting things done without taking the blame.

We do not know how many civil servants, or government staff, there are because the sources, such the Office of National Statistics, Cabinet Office, departmental payrolls and annual reports, all provide different numbers. If the Ibbs principles were followed, staffing would be aligned with policies and objectives, their effectiveness would increase, and personnel management and efficiency would improve. It is odd, to say the least, that, in the 35 years since they were approved, they have not been implemented.

Civil servants are mostly good, hard-working people but they like to operate, and perpetuate themselves, in the environment with which they are familiar.  Benevolently taking a bird out of its cage and helping it fly away, fails; it will fly back into its cage. What looks to us like cruel restriction, looks like protection and safety to them.  Civil servants are no different; structural change and risk are the last things they want. Health Secretary Steve Barclay, speaking at the Spectator Health Summit on 28th November, said “NHS scores the risk of innovation too highly when compared to the risks of the status quo and I think that needs to be recalibrated. This is because innovation tends to be judged, in isolation, in a silo. Take for example the risks around the introducing machine learning. On its own, it may carry some risk. But that risk should be judged against the risk of the status quo, where there may be long delays.”[7] In other words, civil servants should be motivated to recalibrate the perceived risk of change relative to the status quo.

In the civil service, the problem is compounded by change being supposedly in the hands of transient ministers and the avoidance of risk in the hands of senior civil servants.  See Yes Minister for examples. Sir Humphrey always wins.

We need a Prime Minister who recognises the problem and has the strength to implement Ibbs in full.  The government will be the better for it: more focused, efficient, accountable and effective. The way to achieve that is to break the civil service out of its comfort zone and set tough targets for departmental headcounts – maybe a 50% reduction overall. That is not achievable overnight – the pain and cost would be too high but recognising that it is achievable over time would set the government on the right path.









Keep up-to-date with all of our latest publications

Sign Up Here