(UN)REPRESENTATIVE BUREAUCRACY? Are public service leaders unlike the country they serve?

By Colin Talbot

Is the staffing of our government institutions – the civil service and public services – somehow “unrepresentative” of the people they serve?

That they are is a central theme of current UK government thinking. But what precisely do they mean? And is there any evidence they are right?

In a much-reported speech in June 2020 Michael Gove MP, then the Chancellor of the Duchy of Lancaster (in charge of the Cabinet Office and civil and public service reform), set out the case on what is often called “representative bureaucracy” (although he does not use that term).

Gove’s argument was that Government faced huge societal reform challenges but in order to confront them Government had to first reform itself. Drawing on some, frankly sketchy, parallels with FDRs ‘New Deal’ in the USA in the 1930s.

The existing machinery of (federal) Government, Gove claims, was not up to the challenge in its “structure, ambition and organisation”. He goes on to list (somewhat selectively) some of the changes FDR introduced.

Gove says that the UK faces a similar need for change now because of “just how distant, in so many senses, Government is from the people”. Gove does not mean politicians when he speaks about ‘government’ – he is talking much more about the machinery of government and public services, and the public officials who run them.

He goes on to talk about the “concentration of senior jobs” in London. Why, for example, he asks “are so many of those charged with developing our tax and welfare policies still based in London.” 

Gove adds “wouldn’t it be better for those deciding how taxpayers’ money is spent to be living and working alongside those citizens across the country, from Mansfield to Middlesbrough to Merthyr Tydfil”.

“Should we not also be better at recruiting policymakers from the overlooked and hitherto undervalued communities?” Gove added.

In an even more controversial passage Gove spoke about the need for Government to be “less southern, less middle class” and “less anywhere and more somewhere – closer to the 52% who voted Leave”.

These ideas clearly influenced a ‘Declaration on Government Reform’ produced 12 months later (June 2021) and signed by the Prime Minister and the Cabinet Secretary. That document also talked about the need for “re-wiring, and renewal, of government” including the need to “deepen our understanding of citizens in all parts of the country”. 

It goes on to say that government “will look beyond London to all corners of the UK, as part of our mission to be a government more like the country we serve” (emphasis added).


To be a government more like the country we serve” could almost be the definition of a concept known as “representative bureaucracy”. The term first appeared in a book of that title by an American academic, Donald Kingsley, in 1944, about the British Civil Service. Kingsley’s argument was mainly about social class in the context of the extremely upper class domination of the British civil service at the time – and the prospect of a Labour government representing very different class interests.

By the late 1960s – and in the context of the civil rights and women’s movements in the USA – the term had morphed into being about a much wider set of mainly demographic or sociological characteristics. And it is mainly in that context – of wanting a public service bureaucracy that was more “like the country we serve” that debates about representative bureaucracy have continued for more than half a century.

But what is meant by “public bureaucracy”? There is a confusion running through both Michael Gove’s speech and the ‘Declaration’ about what they mean by ‘government’. Does it simply mean the civil service? Or the whole public service? Or does it apply mainly to the popular image of the Whitehall mandarin?


One definition of ‘bureaucracy’ in the UK might be just the Civil Service – currently around 500,000 people. But about 80% of them – 400K – do not work in London. Almost 48K work in Scotland, 36K in Wales and 4K in NI. The other 312K are distributed around the English regions.

The big change in where the bulk of the Civil Service are located already took place, between 1970 and 2010 (see diagram). This was partly due to moving some CS activities out of London and the South East, but it also reflected the growth in CS delivery of things like benefits, employment advice and prisons – all of which are widely distributed.

The wider public service – currently about 5.2 million – is similarly mainly distributed around the UK running the NHS (1.8 million), education (1.5 million), local governments (2 million) and so on. In total 1.1m public servants work in the devolved regions of Scotland, Wales and NI, although not all for the devolved governments. Somewhere close to 3m – 6 out of 10 – public servants work for devolved or local governments.

So while it is partly true that “those charged with developing our tax and welfare policies [are] still based in London” Gove has carefully chosen two examples (tax and benefits) that are still mainly UK-wide central government functions.

(Note: we had to combine London and the South East regions for this graph because early editions of CS Stats had these as a single ‘region’.)

When it comes to “those deciding how taxpayers’ money is spent” locally they are for the most part “living and working alongside those citizens across the country” who use the services concerned. Apart from tax and benefits, the vast majority of people’s interactions with “government” are with local services run mostly by ‘people like them’.

One important caveat to this is that ‘local’ services, in England, are increasingly being run not by locally elected politicians or even local unelected bureaucrats, but from Whitehall departments. In English health, social care and education power increasingly emanates from Whitehall whilst actual local organisation is also increasingly in the hands of private providers.


Lots of work around equalities across all public services suggests that recent decades has seen considerable progress in better representation of women, ethnic minorities, and other previously un-represented groups.

When I looked at this solely for the Civil Service a decade ago (Talbot 2014) it was clear that, demographically and sociologically, significant progress had been made towards a much more representative service.

There is not enough space in this short piece to reproduce all the evidence, but it is fairly clear.


So if the current Government’s attack on the “unrepresentativeness” of public service leadership is not really about geography or sociology, what is it about?

The clue, I would submit, is in Michael Gove’s comment about wanting ‘government’ (i.e. public services leadership) to be “less anywhere and more somewhere – closer to the 52% who voted Leave”.

This is really about a newly Brexiter-dominated Conservative Party wanting to politicise public services leadership. It is their version of Mrs Thatcher’s famous litmus test – “is he one of us?”. An article in The Times in 2019 was headlined “Give public roles to Tories, ministers told.”. And considerable evidence is emerging that this precisely what they are doing.

So is Michael Gove’s populist rhetoric about the unrepresentativeness of public services largely a justification for this program of ‘cultural and political revolution’? It certainly seems the most likely explanation?


Colin Talbot (2014) The British Administrative Elite. The Art of Change without Changing? Revue Française d’administration publique 2014/3-4 (N° 151-152).

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s