It has been a politically tumultuous year. A new Prime Minister, Theresa May, succeeded David Cameron following the referendum decision to leave the European Union. The civil service has continued with the business of government with fewer staff and less money, and is preparing for the challenges of Brexit.

Theresa May has one of the most challenging inheritances of any prime minister in recent times, and much-changed machinery with which to face it. She has already expressed some frustration with Whitehall, and is expected to be more hands-on in how she runs her Government (she chairs half of all Cabinet committees, where she expects a lot of policy development to take place – Cameron chaired only a third).[1] But what does May’s Whitehall look like?

This Whitehall Monitor annual report – our fourth – uses government data to visualise and analyse the size, shape and performance of Whitehall. The decisions taken in Whitehall – shorthand for the administrative centre of British government – have a profound impact on what happens to people all over the country. Understanding the way Whitehall works is therefore vital.

Each chapter considers a different subject: political leadership, the civil service workforce, government finances (and how departments control their spending), legislation, major projects, transparency and accounting for performance.

There are three themes running through the whole report: how well Whitehall is performing, its preparations for Brexit, and openness and the use of data.

How Whitehall is performing

The Government is trying to do too much, although it is continuing to function despite reduced staff and budgets. However, it is extremely difficult to measure government performance.

Civil service staff numbers and budgets have been reduced in nearly all departments…

There are now just under 385,000 civil servants working for the Government. This is down almost 19% since the Spending Review in 2010 (when it was just under 475,000) and the smallest number since the Second World War. Only four departments – the Department for International Development (DfID), the now-abolished Department of Energy and Climate Change (DECC), the Cabinet Office (CO) and the Department for Transport (DfT) – grew in terms of staff numbers between 2010 and 2016. Three – the Department for Work and Pensions (DWP), Department for Environment, Food and Rural Affairs (Defra) and Department for Communities and Local Government (DCLG) – had their numbers reduced by more than a third.

Day-to-day spending budgets have also fallen nearly everywhere. Ring-fencing of international aid, NHS and schools funding means DfID, the Department of Health (DH) and Department for Education (DfE) were three of only five departments whose budgets increased between 2011/12 and 2015/16; DCLG’s dropped by more than 50% and DfT’s by more than 60%.

…but Whitehall, at least, seems to be keeping its head above water.

It’s difficult to measure precisely what impact these reductions have had on Whitehall’s performance, but one important proxy is the Civil Service People Survey; employee engagement is an important indicator of organisational health. In 2016, the survey found that employee engagement had increased in nearly every department, falling only at DH (by 12 points, the largest fall ever recorded by a Whitehall department, which may owe something to a recent round of redundancies).[2]

Broadly, one could say that most of Whitehall is performing reasonably well despite big changes in recent years. Departments continue to function. Ministers continue to receive policy advice. Legislation continues to be passed. Major projects continue to be delivered (although with slightly less confidence than in some previous years). Requests for information continue to be answered. The business of Whitehall continues even after reductions to staff numbers, reductions to budgets, and – in some cases – major changes in what departments do and how they do it.

But this may not be the case for some of the public services that departments are ultimately responsible for. How have budget (and staff) reductions at the Ministry of Justice (MoJ), for example, affected how our prison system operates? What effect have budget reductions at DCLG had on the adult social care system managed by local government? Has ring-fencing the NHS budget maintained quality in hospitals? A forthcoming Institute for Government report, Performance Tracker, looks at five public services – prisons, adult social care, hospitals, schools and the police – and how they have fared over the past few years.[3] All of these services face challenges in the future – and the same can be said of Whitehall departments that continue to be asked to do more with less.

However, the Government is still trying to do too much…

The chief executive of the civil service, John Manzoni, has said that it is ‘doing 30% too much to do it all well’.[4]  Even before the result of the EU referendum, the Government was trying to do too much.[5] Is there any sense of prioritisation?

Although the Government has made welcome progress on reducing the major projects portfolio overseen by the Infrastructure and Projects Authority (IPA), more remains to be done. Crudely, applying a reduction of 30%, as suggested by Manzoni, to the 2015 portfolio would have meant the IPA overseeing 132 projects worth a total of £342bn in 2016; in reality, there are 143 projects worth £405bn. Even these projects represent only part of the Government’s workload, with departments overseeing their own major projects and Government having other priorities – everything from social mobility and ‘just about managing’ families to Heathrow expansion, devolution and public service reform. This remains a challenging to-do list even before considering the impact of Brexit, and of high levels of ministerial turnover – only the Ministry of Defence (MoD) has held on to more than half of its ministers from before the July 2016 reshuffle.[6]

The sense of overload is also borne out by the Single Departmental Plans (SDPs). These were proposed as a ‘single, clear roadmap’, bringing together ‘efficiency, spending round and activity plans’, aligning inputs and outputs, and ‘making clear the trade-offs and choices’ that would allow the civil service to ‘prioritise effectively based on a clear understanding of how our resources can best be deployed’ with ‘no room for “nice to haves”’.[7]

…and has little idea about its own performance.

Unfortunately, the first SDPs published in February 2016 gave no sense of priority. Despite the hard work of civil servants in developing them, ministers appeared to have shoehorned as many of their manifesto commitments into the SDPs as possible: Patrick McLoughlin, then at DfT, had close to 100; Theresa May, then at the Home Office (HO), had more than 60. The Institute for Government criticised SDPs as ‘little more than a laundry list of nice-to-haves’; the Public Accounts Committee has argued that they ‘do not enable taxpayers or Parliament to understand government’s plans and how it is performing’.[8] Only nine departments still in existence appeared to have updated their plans between Brexit and the end of 2016, and no public plans exist for the new Department for Exiting the European Union (DExEU), Department for International Trade (DIT) or Department for Business, Energy and Industrial Strategy (BEIS).

The SDPs show that the way government measures its performance has gone backwards, even from the Coalition’s suite of Departmental Business Plans and impact indicators (which, despite early promise, were difficult to use and were consequently neglected).[9] Without clear, sensible and transparent plans and measures it is impossible for the centre of government, departments themselves, Parliament or the public to understand how well government is doing and hold it to account.

Preparing for Brexit

Creating new departments to manage Brexit proved to be a distraction, but there are some signs that they are beginning to settle in. Some of the departments likely to face the biggest challenges – notably Defra – will do so with much smaller workforces and budgets, and without recent experience of passing a large volume of legislation.

Creating new departments was a distraction, but some are now finding their feet.

Machinery of government changes – creating new departments, and merging or abolishing old ones – always bring risks, in terms of the time and cost involved in setting them up (the creation of DECC in 2008 cost at least £15m) and the distraction they can provide from the actual business of government.[10] This was certainly the case following Theresa May’s decision in July 2016 to create DExEU to coordinate Brexit negotiations; DIT (focusing on how and where the UK will trade post-Brexit); and BEIS, formed from a merger of the Department for Business, Innovation and Skills (BIS) and DECC. Indeed, none of the new departments was able to report its September staff numbers to the Office for National Statistics (ONS).

With three departments – the Foreign and Commonwealth Office (FCO), as well as DExEU and DIT – managing Brexit, time and energy was inevitably wasted in turf wars, fragmentation, incoherence and a lack of clarity around roles and responsibilities. A better option may have been to support a Cabinet minister for Brexit with a unit in the Cabinet Office.[11] However, it appears that the new institutional arrangements are starting to settle down, with civil servants across Whitehall saying that DExEU, at least, has found its feet quickly, working across Whitehall while still setting up its own organisation.[12] These departments do have to get used to the nuts and bolts of government business as well as their Brexit-related tasks; they are already receiving Freedom of Information (FoI) requests, for example, with DExEU missing the 85% target for responding to requests on time in its first quarter of existence.

Some departments will face a heavier load than others and do so with reduced workforces and budgets.

Some of the departments facing the biggest challenges around Brexit have experienced deep staff and budget cuts. Defra, for example, has estimated that a quarter of EU laws (around 1,200) relate to its work, and that 80% of the department’s work is ‘framed’ by EU legislation.[13] But staff levels at Defra have fallen by more than a third – more than one in three staff – since 2010, with a reduction in its day-to-day spending (Resource DEL) budget of more than a fifth since 2011/12. The Home Office – also likely to be affected by Brexit, given its immigration responsibilities – has had its budget reduced by nearly a fifth and a staff reduction of around one in ten.

The Chancellor announced some additional money for the FCO and the new Brexit departments in the Autumn Statement – up to £412m in total for this parliament. It is unclear whether this will be enough; and the Autumn Statement included no details of extra money to help other departments facing their own challenges in withdrawing from the European Union.

Some departments may need to bring in new skills to cope with Brexit. Again, Defra could face a heavier legislative load than before – since the start of the 2010-12 session, it has been responsible for only two government bills becoming acts. Drafting the Great Repeal Bill – which aims to transpose EU law into UK law where practical – is also proving a more complex challenge than expected, which could further add to departments’ workloads. Departments thus face big challenges in planning for and beyond Brexit, many doing so with fewer staff and less money, while needing to carry out relatively unfamiliar tasks.

Data and government openness

The record of the Home Office under Theresa May, the patchy performances of many departments in publishing transparency data, the continued opacity of government finances, and the impossibility of measuring government performance might suggest that government is becoming less open. These issues also highlight its failure to use its own data effectively.

May’s Home Office was one of the worst-performing departments in responding to information requests on time.

Only MoJ and DfE had a worse record than the Home Office under Theresa May in responding to requests for information (parliamentary questions, ministerial correspondence and FoI) on time. It had the second-worst record on parliamentary questions and the worst of all departments on FoI requests. Indeed, its record of responding to 83% of FoI requests on time between 2010 and 2016 falls short of the Information Commissioner’s 85% threshold, below which departments are subject to special monitoring.

This isn’t exactly encouraging for those who had hoped that the new Prime Minister would build on the Coalition’s open government initiatives.[14] (Nor was the fact that the details of her first Cabinet committees were published only after an FoI request from the Institute for Government).[15]

Departments are withholding information in response to more FoI requests than in 2010, and performance is patchy on transparency releases.

Fears that the 2015-16 Independent Commission on Freedom of Information might threaten the transparency provided by the Act turned out to be unfounded.[16] But government has become less transparent in its responses to FoI requests over the past six years: at the start of the Coalition, government departments withheld information in full in response to two out of every eight requests (25%); by the second quarter of 2016, this had risen to two in every five (40%). Theresa May’s Home Office also went from 25% fully withheld to 40% over the same period.

Freedom of Information is only one aspect of the wider open government agenda. The Coalition aimed to be ‘the most transparent and accountable government in the world’, and David Cameron wrote to departments shortly after becoming prime minister about transparency and opening up data.[17] Analysis of two of the transparency releases mandated by that letter – departmental monthly spend over £25,000, and departmental organograms – shows that publication has been patchy across departments.

Departments are supposed to publish their monthly spend by the end of the following month. But – even allowing a few days’ grace – 51% were published late (and 3% not published at all). Infamously, the Cabinet Office – the department responsible for open data policy – was over a year behind in publishing some of its spend data (and did so only after repeated FoI requests from the open data start-up Spend Network).[18] No spend has yet been published by DExEU, DIT or BEIS. Even where departments have published data, there are real problems with using it: our previous work with Spend Network has shown that some data is not recorded, there is a lack of clarity about how government organisations are defined, and it can be difficult to work out which private companies are ultimately benefiting from government contracts.[19]

Departments have also been inconsistent in publishing their organograms, which include details of departmental units, salaries, grades and lines of management (interestingly, HMRC – HM Revenue and Customs – and DH are among the best at supplying both organograms and spend data). Recent improvements by the team at, including making it easier for departments to upload data, are welcom.[20] But again, shortcomings in the quality of the data – for example, failure to record changes over time in the names of units within departments, and inconsistencies between the organograms and other datasets, such as the ONS on the professions of civil servants[21] – suggest that departments are not using what could be a useful workforce management tool.

It is still more difficult than it should be to track government finances and government performance.

Increased government openness over the past few years has made it possible to compile this report. But there are still difficulties for the public, Parliament and presumably the Government itself in understanding how government operates. For example, in the documents they publish and lay before Parliament, not all departments explain clearly why their spending plans have changed, with the Treasury (HMT) one of the worst performers. Different financial datasets use different definitions – in some, HMRC and HMT are separate; in others they are combined – and, as our forthcoming Performance Tracker report will show, tracing government spending becomes even more difficult when looking at individual public services, such as the NHS. As with finance, so too with performance: the shift from Departmental Business Plans under the Coalition to Single Departmental Plans under the Conservative Government makes it nearly impossible to understand how departments are performing on their policy priorities.

Even where government does release data, there are issues with how it is published (in PDF format, rather than more usable spreadsheets) and with quality (everything from how departments are defined, which can vary between datasets and make comparisons difficult, to the different professions civil servants belong to, where we don’t know the answer for one in ten staff). It may be that staff reductions and recruitment freezes are affecting the ability of civil servants to do this work, although the creation of the new digital, data and technology profession may help.

But one overriding conclusion from the patchy publication and variable quality of data has to be that departments themselves cannot be making enough use of their own data, known as ‘dogfooding’; if they were, the data would be better and more easily accessible. If this is the case, then government is not using data – on everything from its size and shape to its performance – that could mean better outcomes for departments and for citizens.

Over the past couple of years, open data has lost a number of champions in government, on both the political side (e.g. Cameron and his former minister for the Cabinet Office, Francis Maude) and the administrative (e.g. Paul Maltby, former head of data at the Government Digital Service), and some have detected a sense of drift.[22] The failure of the UK to send a minister to the most recent Open Government Partnership summit in December 2016 is a worrying sign.[23] But good work continues – for example, at Open Defra, or the Register Design Authority in the Government Digital Service.[24] The forthcoming (if much-delayed) Digital Transformation Strategy is expected to create the role of chief data officer.[25] There are strong communities, within and outside government, championing data and openness. Those communities need to show why opening up and measuring up can have administrative and political benefits; government needs to realise those benefits.


Download PDF download pdf