I received the following comment from a serving civil servant who wishes to remain anonymous. I publish it here (with their consent) and add a comment of my own at the end:
Although a civil servant I have some sympathy with Margaret Hodge in the recent debates over accountability; although the principle of civil servants being accountable to Parliament only through Ministers is fine, there’s a good case that it’s not working in practice. But it seems to me that there are some very significant implications of such a change which have not really been acknowledged.
At the moment, one could argue (not entirely fairly) that civil servants have power without responsibility when it comes to delivery, and Ministers vice versa. The risk is that we reverse that. I have no inside knowledge of what happened, for example, in relation to the ill-fated fire service reform programme, but it seems likely that one of two things happened:
– civil servants were confident it was deliverable, and advised Ministers as such, but failed to deliver
– civil servants advised that delivery was highly risky, but Ministers decided to go ahead anyway (and as good civil servants they did their best, muttering ‘on your head be it, Minister’)
But the dynamic in the second case changes fundamentally if civil servants are to be publicly responsible for failure to deliver. It would be unreasonable to put civil servants in a position where they were held publicly to account for failure to deliver a project which they knew was highly risky and possibly undeliverable. In the one case where civil servants are publicly accountable – as Accounting Officers – there is of course a get-out clause, the accounting officer direction, and if public accountability were to be extended it seems to me that a similar mechanism would need to be put in place, allowing a civil servant to say to their Minister ‘I do not think this plan will work and I cannot be held responsible if it doesn’t, but I’ll do my best if you really want to go ahead’. That, of course, would change the nature of the relationship between Ministers and civil servants: it may increase civil service accountability but it would also increase their power. No Minister would be happy to be told publicly by their civil servants that their pet scheme was unworkable, and they may feel their opportunity, as a democratically elected politician, to take difficult decisions was being restricted by the bureaucrats.
Such a change might improve delivery of public services, of course, but it might equally lead to risk-aversion: if my Minister wanted to introduce a highly risky project they thought would transform society, as a civil servant I might be reluctant to go along with it for fear it would go wrong, destroying my reputation and potentially my career for something I may not agree with: there would need to be some thought given to incentives and recognition of risk-taking amongst civil servants if it were to work. That leads to my other thought, which is that accountability cannot be based on the principle of ‘heads Ministers win, tails civil servants lose’: accountability cannot simply be in the form of criticism when something goes wrong, and there should be clear opportunities for civil servants delivering challenging projects well (it does happen!) to be publicly recognised for that.
This contribution raises a difficult issue – the boundary between ‘political’ or ‘policy’ accountability and that of implementation or ‘delivery’ (as the previous government would have called it.
Some policies fail because they were ill-conceived by Ministers. At the moment Ministers can get away with blaming ‘implementation’ failures. Indeed, after the ‘Derek Lewis’ affair the then Home Secretary Michael Howard used precisely this defence – he was accountable only for ‘policy’ and Mr Lewis, as DSG of the Prison Service and a civil servant, was responsible for delivery. There was even an attempt following this codify this as Ministers were ‘accountable’ but not necessarily ‘responsible’.
The problem was, in the Prison’s case, that it was also about policy failure. Mr Howard had set or approved the Prison Service’s Key Performance Targets, its Framework Document, its Business Plans, etc. all of which failed to stress the importance of keeping category A (high security) prisoners locked up. All of these were changed after the event, but Mr Lewis was managing delivery in a faulty framework laid down by, or at the very least approved by, the Secretary of State.
The problem in accountability terms is where do we draw the line between ‘policy’ (Ministers) and ‘delivery’ (Civil Servants)? There is no simple answer. But the current position is absurd. Ministers can claim ‘accountability’ when it suits them and blame ‘responsibility’ when it doesn’t. In practice, no-one can hold Civil servants properly to account. Our system is hopelessly out of balance and requires greater transparency and accountability of Civil Servants to Parliament.
As to ‘how can it be done?’ (a last ditch defence from the FDA and others) the answer is simple: it is done in plenty of other countries without civilisation collapsing.