Civil Service Impartiality — When is an elephant in fact a windmill?

I came across a blog recently about civil service impartiality that has had me thinking. I missed it when it was originally published last year, but its author — former DfE civil servant and SpAd, Iain Mansfield — has been resharing and it caught my eye this time.

Civil Service impartiality is something I care deeply about. Although I don’t write very often on this site, when I do, issues around the operation of the civil service are a common theme. And I respect Iain’s knowledge and views in this area. After all, if somebody who has moved from being a politically neutral civil servant to becoming a Special Adviser (perhaps the most overtly political role in the Department) cannot shed some light on how political neutrality plays out, then who can?

Overall I think Iain makes some excellent points about the nature of being a civil servant in Britain today, and it helps that we both have experience of the same department — the Department for Education — because I recognise the things he discusses. But there are a number of ways in which I think his blog is giving only a partial description of circumstances and, as is always the case, the voice that is missing to provide alternative views is that of the civil servant. Because of the very political neutrality that Iain (and I) are worried is being eroded, civil servants do not comment on such matters.

So, as an ex-civil servant like Iain, I share some thoughts on the problems he raises — what he describes as the five elephants.

1 — Poor Performance

The blog makes the entirely legitimate point that some civil servants are not very good at their jobs. This is of course true — it is true in every organisation, in every walk of life, at every time in history. It describes a number of examples of projects that have been badly delivered or delayed and implies that the fault is that of the civil service. It specifically says:

It is worth saying that neither ministers nor SpAds are typically experienced procurement experts, IT specialists or project delivery managers — and nor should we expect them to be. It is reasonable to expect the large civil service organisation, with its many fairly highly paid people, to be able to deliver this.

I don’t disagree with him on principle, but there are a number of rather important caveats that are ignored in this statement.

First, it assumes that the thing being asked to be delivered is possible in the first place. He cites the example of the National Tutoring Service as an example of a failure of procurement, which is undeniably true, but it was also rolled out at an unwise pace, with too few resources. The failures in procurement were largely failures of policy design — you can, after all, only procure what you are instructed to — and these failures of policy design rest squarely on the shoulders of ministers.

Second, I think it is implicit that if you are not the expert but you are surrounded by experts, then you should assume you listen to them. Procurement, IT, programme and project delivery — all of these take experience and training to do well and if you are head of an organisation that has people with such training and experience, taking their advice would seem to be sensible. In other words, if ministers took all the advice offered on the National Tutoring Service, accepted it fully, and then it failed, the blame would rest firmly on the civil servants offering that advice. But that is not what happened in the case of the National Tutoring Service or indeed most (every?) other project.

If there is a sense that civil servants drag their feet, take too long to do anything, and are trying to undermine policy ideas — that is in fact part of the nub of the blog— then good advice can be ignored and — perhaps more worrying — the civil service begins to respond by only offering palatable advice. ‘Serving the Government of the Day’ becomes telling ministers what they want to hear rather than ‘speaking truth to power’. A vicious circle is created where the quality of advice declines, and trust in advice declines… so actions happen that lead to a further decline in the quality of advice and so on.

Third, it assumes that the civil service has the resources needed to deliver the policies decided. It is easy to point to the number of civil servants, to elide into conversations about pay (inevitably moving on to ‘gold-plated pensions’) and say they should just get on with the job. But the harsh truth is that there are seldom enough people to deliver the work. In my own time in the DfE, I ran a division with a notional headcount of over 100 people — that is even within the limited resources of the civil service, it was agreed that I needed more than 100 people to deliver the complex, major project I was leading. But I routinely ran at a vacancy rate of over 20%. At no point was that ever an acceptable reason for not meeting a milestone or getting a project delivered. So — in such circumstances — it is inevitable that quality suffers.

Perhaps the most worrying element of the blog in this section is the comments on Dominic Raab, who was of course found guilty of bullying his staff:

There is a disconnect between Ministers being accountable for performance, but having no ability to transfer, fire or reward those responsible. It is no good to say that the best Ministers can manage under these constraints — yes, they can, but no private sector organisation would dream of tying their leaders’ hands behind their back in this way. We can see this at the heart of the Raab affair: whether or not you agree he was bullying, it appears to be undisputed that one element was that there was a senior civil servant who he did not believe was delivering on an important task, who he then reprimanded.

First, Dominic Raab was a bully. That was found to be the case through an independent process. There are nuances and mitigations (as there are in all things in life) but to cast any doubt on that fact (“whether or not you agree this was bullying”) serves to undermine any case here about civil service performance by giving space to the idea that it is okay to bully a civil servant if you think they are not doing their job well. I hope that most people can see that this is not okay.

Second, it is true to say that no private sector organisation would try to constrain a CEO in terms of staffing decisions (although they absolutely would in terms of bullying) and that CEOs can — within the law — look to make people redundant or change their roles. It is also true to say that no private sector organisation would select a CEO based on the sorts of political calculations that feature in the choosing of a Secretary of State, nor that the performance of many Secretaries of State (of all political parties) would be tolerated for long in a private sector organisation. The argument that the electorate are the shareholders is a fatuous one (no elector votes on who gets to be a minister in a particular role, and the five-yearly cycle of general elections is no check at all on the performance of a Secretary of State). So, if you want to start arguing for private sector ways of operating in the civil service, you need to argue for private sector ways of working in the ministerial team too. They are symbiotic — you cannot change one without the other.

Where I do agree with Iain is the need to look at the performance management of the civil service. For all I have added above, it is true that some civil servants are not very good at their jobs — as I say is the case in all walks of life — and it can be too difficult to fire them. It is, as it happens, far easier to move people around that perhaps Iain implies. In fact, the ease with which people can be moved around the system is one of the reasons why genuinely poor performance is not tackled. I am not against his specific suggestion that ministers might, formally and rarely, ask for an official to be moved. It does happen informally, and there are negative consequences if there is any sense that the reason is because the advice was unpalatable, but I think the idea is worth exploring.

2 — Recognising certain issues are politically contentious and 3 — The embedding of campaigning, activist groups in the public sector

I am taking these two points together as I think they are so closely connected, as I will try to explain below.

Of course the civil service knows that ideas are politically contentious. That is the bread and butter work of so many civil servants, especially those working around ministers. What this section of the blog is really about is diversity. I know this is something Iain feels strongly about because during the time we overlapped at the DfE (and I should say that we never really had anything directly to do with one another), I know he tried to stop some of the ways of working that he criticises the civil service for in his blog.

I start from the premise that the main purpose of the civil service is to enact the policy of the government of the day. In order to do that, it is incumbent on the leaders of the civil service to ensure it has the knowledge, capabilities, and capacity to respond to whatever a particular government chooses to do. This by necessity involves long-term planning around issues such as recruitment, HR, and ways of working, and I think it is important that such planning is determined solely by what will be most effective in securing the workforce needed to be effective civil servants (that is to understand policy, draft legislation, deliver major projects), not shaped by politics. That means that such policies will remain constant even as political tides shift — from the world view of a Justine Greening to that of a Kemi Badenoch, and from one political party in power to another. Ministers may have particularly strong views on gender- neutral toilets, for example, but this should be irrelevant to the leadership of the civil service unless the government chooses to enact specific laws about such issues.

Iain writes:

Yet in the civil service, very one-sided positions on these are fully embedded within training courses, HR policies, within ‘staff support groups’ or in statements made by senior leaders. Sometimes these directly contradict the policies or positions that have been set out by Ministers. It should be as unacceptable for a staff member to write an intranet post calling for gender self-ID, or about ‘white privilege’, as it would be for them to write a post calling for the nationalisation of the railways or the abolition of inheritance tax. Not because any of these positions are necessarily wrong — but because they are contested and political. Similarly, email signatures and lagnards [sic]with rainbows, BLM hashtags or pronouns in signatures, or other public statements showing support for political causes badly compromise the appearance of impartiality.

As I have already set out above, if any such HR approach contradicts the position of a minister, then tough. The civil service cannot make long term decisions about HR policies based on the views of any individual minister, not least because these are not consistent even within the governing party, and the high level of ministerial turnover means trying to respond to every personal view would lead to chaos. Instead, the civil service should be focused on how it can bring in the requisite skills to do its job and the statistics on how it is performing in this way are not positive. The civil service has thus far failed to be an organisation that is able to recruit and promote talent equally from the whole of society, especially into positions of senior leadership.

The Government’s own strategy — the Civil Service Diversity and Inclusion Strategy 2022–25 (which for completeness includes an endorsement from the relevant government minister at the time, Nigel Adams) — includes the following objectives:

To enable a Civil Service that:

• understands and draws from the communities it serves — drawing from a range of backgrounds, experiences and locations

• is visible to everyone — engaging the communities we serve and showcasing what the Civil Service offers

• is flexible — supporting innovation, performance and engagement

• welcomes talent from wherever it comes — attracting the best talent from all backgrounds

Overall, much of the data on diversity within the civil service is improving (although not necessarily at a fast enough rate for some), which rather suggests that the approaches that the blog derides might just be working. Again, to be clear this isn’t about diversity as an end in itself — it is about ensuring that the civil service is able to recruit the best talent regardless of background and — given the size of the civil service and therefore the data sets we are talking about — if we are to accept that any particular group being under-represented proportionately is okay, it means accepting that group is less likely to have people of the required talent. I hope, once again, we can agree that is not a positive position.

Which takes me on to the groups the civil service is working with — groups such as Stonewall. Such organisations are routinely engaged by the civil service to help address barriers to the recruitment and progression of staff from certain backgrounds — in Stonewall’s case the LGBTQ+ community — again as part of a broader point about ensuring that the civil service is able to recruit and promote the most talented people. Some civil servants may agree with their wider views but that isn’t why they are selected. They are chosen as partners for their track record in supporting the recruitment and development of staff from certain backgrounds. At times their broader perspective may be in line with government thinking, and at other times they might not. That just should not be a relevant factor.

I also think the blog tends towards the ridiculous in a few ways. Is he really suggesting that rainbow lanyards should be banned, for example? For the last period of my time in the DfE I used a lanyard that was created to celebrate the centenary of women getting the vote — a rather natty looking gold affair, with suffrage purple, white, and green stripes upon it. It was created by the Government Equalities Office and the relevant minister at the time wore one. After I started wearing it, I largely forgot about it. Should I have paid more attention? It is definitely a political statement — votes for women is an intensely political decision, albeit thankfully not one that has much support for the contrary view any more. What is the statute of limitations on such ideas? Many civil servants wearing rainbow lanyards will have been given them at events (they were handed out liberally in my time) — and this is partly the blog’s point: that the civil service promotes inclusion. But, as I have said above, this is a calculation about creating a modern and effective working environment, not a political statement. Should these civil servants stop wearing them? Should civil servants tell serving ministers that they cannot let officials wear lanyards that promote a government policy just agreed?

The same is true of a he/him type statement at the bottom of an email signature. There is no government policy on this. There is no legislation preventing it or guidance discouraging it. There is also no requirement to do it. It is a personal choice by the individual to help others understand their identity. Are we really saying that is a political decision, a contentious decision that people should shy away from? To be clear, if we are and therefore we are asking that civil servants refrain from using he/him statements at the bottom of their emails, then we are explicitly moving from a position of neutrality, to one where we are clearly taking a side in the political debate. That is the opposite of what we should be doing.

I want to talk a little about the specific point the blog makes about the Black Lives Matter slogan. I was a senior civil servant when George Floyd was murdered in the US. I’ll admit I was taken aback by the way it made those who were black and who worked for me feel. It happened in the US, not here, and our history of race relations is different in many ways, but that moment was shocking and it had a genuine impact on people in my team. I could, of course, have decided that was nothing to do with work — that it was personal and people should leave it at the door — but I don’t think that is an effective way to manage people, nor to get the best out of them. Moreover, I am certain that other circumstances would not be treated that way. When the Queen passed, mourning was observed in all government departments, the Permanent Secretary of the DfE was amongst those paying respects on behalf of the department. We accepted that was an external event that had a profound impact on people, and rightly so. My point is that we do not get to decide what has an impact on people. We either listen and accept or we don’t — we can’t pick and chose based on what we might believe is important. As it happens, I never came across anybody who had the Black Lives Matter slogan in their email signature (I am not saying it didn’t happen — I just didn’t see it), and I agree it shouldn’t be there in the same way that any external organisation shouldn’t be on an official signature. But I do think it is legitimate and important to discuss the issues that affect staff, whether that is the Black Lives Matter movement, the murder of Sarah Everard, or broader issues such as the menopause.

4 — Group think

Group think is a serious risk in any organisation. Although the civil service is numerically large, the number actually working directly in advising ministers and shaping policy (the senior civil servants and those in more junior but direct policy roles) is much smaller. The definition is that a group of individuals begin to value conformity over critical thinking, who spend time reinforcing rather than challenging one another's ideas. As Iain defines it:

We saw this very visibily [sic] on Brexit, where the vast majority of senior civil servants and Permanent Secretaries supported Remain. We see it on issues related to equality law — as discussed above — or in the DfE, at the time when Theresa May tried to reintroduce grammar schools. We see it when the civil service union threatens to go on strike if told to implement the Rwanda plan.

This paragraph is worth unpacking. First, I don’t think we know what the vast majority of senior civil servants and permanent secretaries supported in terms of Brexit. There has been some reporting of individuals, but I don’t know of anything comprehensive, and I think this likely stems more from the broader (and I think incorrect) arguments that are put forward amongst Brexiteers that the civil service has resisted Brexit. This excellent Institute of Government blog unravels that argument. Second, it is worth being clear that remaining in the EU was the official position of the government of the day (albeit one that was undermined by the decision by Cameron to disapply collective cabinet responsibility for the referendum campaign on that issue). So supporting the government of the day meant supporting the position to remain, no? As it happens, the civil service took a much more neutral stance, and rather than get involved in the unedifying spectacle of the Brexit referendum campaigns, instead got on with the business of responding to what ministers asked of it.

On the issue of equality law, I can find no example of the civil service doing anything other than advising ministers on the basis of the existing law. That law might not chime with the views of a particular minister, but it does not stop it being the law.

On grammar schools, teams of civil servants were ready to take forward action to change rules around the creation of new grammars. Civil servants have been involved in any number of structural changes in the school system, including the introduction and expansion of MATs most recently. What I suspect might be behind Iain’s concern is two things: first, that the grammar school policy was framed in terms of social mobility when the evidence suggests they do little to help that and it is a duty of civil servants to point that out, and second, that the policy failed for political reasons, not least that Teresa May’s Secretary of State at the time, Justine Greening, offered lukewarm support for new grammars at best. On that latter point, who does Iain want DfE civil servants to serve? Justine Greening or Nick Timothy?

The Rwanda plan is also a questionable example. The main issue cited by civil servants is not that they agree or disagree with the policy per se, it is that they object to being required to do something unlawful. Until the government is able to establish a clear and unambiguous position that deportations to Rwanda are lawful, any move to take it forward places civil servants in an intolerable position. The civil service code requires them to not do anything unlawful, even if asked by a minister. The ‘I was only following orders’ defence is no defence at all.

But what is more important is that none of the examples cited are in fact examples of groupthink. In fact, I posit that the belief in Brexit despite the evidence, the belief in grammar schools despite the evidence, the belief in a Rwanda scheme despite the evidence offer better examples of groupthink — that is the valuing of conformity to a particular policy line over critical thinking — than anything the blog cites. In fact, one of the irony’s of the blog is that the way to get more diverse opinions is to ensure that the civil service is promoting talent from a range of backgrounds and views, which leads to the sorts of HR processes that are questioned. He appears to be saying that the civil service is systemically recruiting and promoting a narrow band of people (left-leaning, well educated). I think many would agree and the data back that up (at least in terms of education level, as well as in other ways such as race). So we agree that there is a structural and systemic recruitment and promotion issue within the civil service, and that addressing that will lead to a greater diversity of thinking, which counters the risk of groupthink.

5. The use of accusations to bring down those with opposing political views

This seems a pretty thin section to me. First, we are talking about isolated examples, with reporting in the Telegraph cited which specifically notes that the actions of the civil servant were not followed through. So it is an example where any attempt by the civil servant to bring somebody down (and I don’t know the full particulars) in fact failed. So the system worked?

He also notes a specific example he witnessed where a government advisor on one policy was questioned in terms of suitability because of his views on trans rights. I don’t want to name the name (as Iain hasn’t) although I suspect I know who he is referring to. It is worth noting two things.

First, it is literally the job of civil servants to highlight anything — anything — that might bring the Department into disrepute and that includes politically controversial statements of whatever kind. For example, following the blog’s previous points, if somebody was a strong supporter of the Black Lives Matter movement, this would also likely be shared as something to consider even if it is irrelevant to the role they are being considered for. It doesn’t stop ministers making a decision — it didn’t in the case Iain notes. And it hardly amounts to an attempt to oust anybody! It is literally officials doing their job.

Second, there are countless examples I saw where people who were eminently qualified to provide advice to the Department, to act in particular roles, were discounted because their views didn’t align with those of ministers, including on issues unrelated to the role being considered. Officials had to get increasingly good at looking at all possible social media channels because one minister in particular would spend a lot of time doing that as well, and rejecting possible people to work with because of something (lawful, measured, and not abusive) said in a tweet some years before. You didn’t want to be the official who missed a potentially contentious tweet (at least in the eyes of the minister). In fact, rather than attack the civil service for highlighting the potential issue with the person Iain refers to, I applaud them for doing their job and raising the issue without fear or favour knowing they were likely to get short shrift from ministers and SpAds. I worry more that officials do not suggest well qualified people whose views on other issues do not align with ministers than that there is some sort of progressive bias.

In fact, returning to the question of whether Stonewall is an appropriate organisation to work with, by the blog’s logic in this section, the sole criterion for determining that should be the extent to which engagement with Stonewall has met the terms of its contract (so around the recruitment and progression from LGBTQ+ backgrounds) and not based on any other statements, a long as they are lawful, regardless of whether they align with minister’s thinking.

Some concluding thoughts

I think there is merit in ensuring that the civil service understands the ideology and thinking behind the government it serves and I think that a reading list of prominent thinkers that have influenced ministers is a good idea. It would be stronger if we could be sure of ideological consistency between ministers of the same party and of a lower rate of turnover of ministers, but nonetheless I think a reading list has merit.

I also think the last 14 years of Coalition and then Conservative government has seen the civil service rise to the occasion to deliver government policies and priorities extremely well. The failures in government policy making more recently are not to do with some nebulous, progressive, civil service plot but rather because government has too often been politically unstable, ideologically unclear, and uninterested in long-term policy delivery for some time. Civil servants have to point out evidence and information to help ministers decide. When they decide, civil servants take forward those decisions — within the bounds of the law. Nothing I can see in the blog deviates from that.

To be clear, it is not that I don’t think the civil service needs reform, and that issues around politicisation are not real and of concern. It is just that I think the areas Iain highlights are largely misleading, and take us away from a more fruitful conversation about how to reform the civil service to make it fit for the future. These are not the five elephants in the room that nobody is talking about. They are windmills being tilted at as a distraction.