Features: May 19th, 2006

The Future of the Civil Service – A Growing Debate

By Colin Talbot

This article was first published in Public Management and Policy and is reproduced by permission of the Association. http://www.cipfa.org.uk/pmpa/index.cfm

I should perhaps start this article with my qualifications for discussing this matter. First, I have some experience of working as a public (but not civil) servant, mainly in local government. Second, I have spent an awful lot of time researching and observing the civil service. Third, I have also spent a lot of time acting as a consultant to various civil and public service bodies.

The contrasting roles of researcher and consultant are especially interesting. The first involves a rigorous search for the truth, usually with some small, patchy and uncertain success. The second involves, usually, simply being told the truth. I have always found that people who are paying you to help sort out problems tend to be a lot more forthcoming about what is really going on than people who are wearily responding to yet another researcher’s inane questions.

Unfortunately, usually having been told the truth as a consultant, it is a bit tricky getting that knowledge into the public domain, for obvious reasons. But there are ways.

Over the past few months there has been a flurry of activity and speculation about the future of the civil service. Parliamentary committees and think-tanks have launched inquiries. Two draft civil service bills have been published. Issues about civil service reform were, probably for the first time ever, a central part of a general election debate with the names of the main political parties’ respective efficiency champions—Gershon and James— being chucked around as if they were premier league football stars. And the appointment of Sir Gus O’Donnell as head of the civil service has been surrounded by another flutter of speculation about what it means for the future of the service.

This article looks at the scale and scope of the issue when we start to talk about the future of the civil service: is there a problem and if so what is its nature? The article reviews, briefly, some of the reforms which have been tried over the past 25 years or so, and discusses whether any of them ‘worked’? Finally, I will turn to Lenin’s famous question: ‘what is to be done?’

What is the Civil Service?

Much debate about the civil service is often not overly informed by an understanding of what it is and what it does. When most commentators talk about the ‘civil service’, what they really mean is the relatively small proportion that work in the hub of government, the ‘core executive’ as it has become known in the academic literature or the rather more colourful ‘Whitehall village’ as it was dubbed 30 years ago by Hugh Heclo and Aaron Wildavsky.

The half-million plus people who work for the civil service are not all Whitehall villagers. The vast majority—probably over 90%—are actually engaged in what we now call ‘service delivery’: collecting taxes, paying benefits, issuing driving licences, supervising abattoirs, running prisons. But 90% of public services are not delivered by civil servants. The civil service as a whole forms only about 10% of the overall public service in the UK, so the vast bulk of ‘service delivery’ is actually outside of Whitehall’s direct control and management.

Which functions are inside ‘Whitehall’ and which are town-hall, or ‘no-hall’, is largely a matter of historical accident. In Greece, roadsweepers are civil servants and, in Germany, teachers. In Denmark, most taxes are collected and most benefits paid out by local government. On a usually small scale, functions move around in the UK— abattoir inspectors used to be local government employees and are now civil servants. There is absolutely no rational or absolute way of deciding which service delivery functions must be part of central government and which not. It should be added that even that most prized of civil service functions—policy advice to ministers—is rather more privatized in some other countries than it is in the UK.

The relationship between the civil service and public services as a whole is a complex one. Some services are managed directly within ministries— for example the immigration services. Some are managed at arm’s-length within the civil service: executive agencies. Some are managed at slightly greater arm’s-length through non-departmental public bodies (NDPBs) or, as they are usually (and mistakenly) called, ‘quangos’. At national level, there are some public services which are even ‘further out’ and have public corporation or other status, for example the BBC or the Bank of England. And, finally, at the national level there is also that great anomaly, the National Health Service (NHS). At local level, there are a number of non-democratic localized services (for example the police) as well as those controlled by local government. I will leave aside the further complications introduced by devolution in Scotland and Wales.

The point here is a simple one: Whitehall has an amazingly complex array of relationships with the public sector as a whole. There are few, if any, private corporations which have such a wide range of functions and, more importantly, such a diverse set of management relationships. Most studies of ‘strategic style’ in multifunction corporations suggest that they tend to adopt a single approach to managing their subsidiaries—for the civil service that is not even an option.

Whitehall does not, of course, just manage public services, whether directly, at arm’s-length or indirectly. It is responsible for advising ministers on, and then implementing, public policies. These often range far wider than just public services and can involve a whole range of private and third sector actors.

Does the Civil Service Work?

There are many extremely talented and dedicated civil servants; the majority are hard-working, committed to an ideal of a public service ethic and a good number are very clever indeed. Whitehall also has its share of problem people. What worries me is not a few problematic individuals or a lot of very excellent ones, but a system which sometimes make them all seem collectively like Whitehall village idiots. If that sounds harsh or extreme, I suggest a short audit of the failed policies which have cost this country billions and, more importantly, sapped the credibility of the public space in which the civil service has been at least complicit.

Taking service delivery by the civil service itself first. The picture is at best patchy. There have been some notable problem areas: child support, prisons, prosecution services, some areas of tax collection, agricultural support etc. There have also been some notable improvements, especially since the introduction of executive agencies from 1988 onwards. Even here, however, the picture has sometimes been problematic—the Passport Agency crisis for example.

When it comes to the civil service’s role in relation to the services which it effectively or largely ‘manages’, the picture is even more spotty. Health, education, criminal justice and transport have seen a series of chronic problems—some policy related but also many much more managerial in nature and in which the civil service often plays a crucial role: failed IT projects, procurement fiascos, or botched building schemes. These have been catalogued in numerous National Audit Office (NAO) and select committee reports. Initiative overload, too much topdown instruction and constant chopping and changing of priorities has been a constant refrain from front-line service providers.

If there is any doubt about managerial problems in the current civil service it is removed by two words: Gershon and Lyons.

Since 1998, at least, we have supposedly had a management system for the civil service which includes resource accounting and budgeting, comprehensive spending reviews, public service agreements (and, for a time, service delivery agreements) and departmental planning processes. On top of this, there has been a whole battery of units at the centre promoting reform.

And yet we find that, despite all these systems and incentives, the civil service has managed to locate in the most expensive places, run up staffing numbers and run down efficiency to the point where the most ambitious set of efficiency reforms ever attempted—by an order of magnitude larger than any previous effort—is required to put things right.

What about policy? Surely the civil service is good, nay excellent, at that? The litany of policy disasters of the past couple of decades suggests that, at the very least, this is a question for discussion: poll tax, BSE, arms to Iraq, arms in Iraq, foot and mouth, the channel tunnel rail-link and so on. Shortly after the 1997 election, the then head of the service declared that the civil service rarely bothered to go back and see whether a policy which had been pushed through a few years ago was working. While this has changed a bit, it is still problematic. And there has been a series of initiatives aimed at improving policy making—the illfated Centre for Management and Policy Studies, the Policy Hub, the Strategy Unit, the Magenta Book and, most recently, the move to create a policy profession—all of which suggest that all is not quite as good as it could be.

Ministry of Excuses

Of course on policy-making the civil service has always had the perfect alibi: ministers decide. If a policy fails, chronically or spectacularly, it is not their fault constitutionally or in any practical sense because under our arrangements only ministers are held to account. Even in areas of service delivery and management this is increasingly the case as the shift to what might be called ‘public management policy-making’ of the past 20 years means that here to ‘ministers decide’, not mandarins.

Some continue to assert that our civil service is ‘neutral’ or ‘impartial’ and as one critic has recently written ‘impartiality doesn’t lead to a passion for public service reform’. This is dead wrong. Our civil servants are ‘serial monogamists’—wedded body and soul to the government of the day. They may ‘speak truth unto power’, but only behind tightly shut doors and not too frequently then. I may, of course, be wrong about this, but we have no way of knowing for sure and civil servants, under current arrangements, have no way of telling us. The evidence that does exist— accounts of specific policy disasters such as the poll tax (see Failure in British Government by David Butler, Andrew Adonis and Tony Travers), arms to Iraq (Scott Inquiry), arms in Iraq (Butler Inquiry) and the Derek Lewis affair—suggests that the relationship between senior civil servants and ministers is far too symbiotic to be healthy.

Under the quaintly named ‘Osmotherly Rules’, for example, civil servants giving evidence to parliament do so only as spokespersons for their minister because they have no ‘constitutional personality’ separate from their ministers and it is ministers who are called to account.

It is true that, in the past, civil servants often developed a sort of semi-detached air, but this was a result of the problems of having to change direction every time there was a change of government, or minister, or when they moved jobs. It is no accident that cricketing metaphors, with all their resonances of gentleman and players and playing the game, are so current within the mandarinate. They suit the serial monogamist role perfectly.

There is one small exception to this: the role of some agency chief executives. Since around the mid 1990s some of the more independent-minded of these people have spoken out, both publicly and in parliament, clearly not as mouthpieces for their ministers. And some of them—notably in prisons and child support—have also been held accountable for failures directly, as opposed to through their ministers. In this limited field we have seen the emergence of what might be called dual-accountability: to the executive and to parliament. This position would be fairly familiar to observers of many other democracies, where their civil service equivalent is usually not seen as the sole plaything of the executive.

Government and Public Services

There is a fundamental tension in modern democracies about the role of public services. They have to be democratically accountable to the executive government of the day. This government often represents only a proportion (in our system not usually even a majority) of citizens. At the same time they are public services and must act impartially to all citizens and be seen to be ‘public’ services. Despite those who want to be able to declare ‘not in my name’ when government does something they don’t like, public services have to be able to credibly act on behalf of all, while being directed only by some. This fundamental tension is handled differently in different systems, but in most cases there is some more or less explicit ‘separation of powers’, which is meant to provide a means of balancing these two important but contradictory values. In most cases the civil service, for example, is accountable to both executive and legislature and, through establishment in statute, to the courts as well. This triple accountability is meant to provide a check on unbridled executive power and enable citizens to feel comfortable that the civil and public services are ‘theirs’, even if they did not vote for the current government. It is a difficult balancing act.

In the UK’s case, I would argue, the balance has tipped almost totally to the executive and the main balancing mechanism—parliament— is very weak. The courts have, to a very limited extent, stepped into this situation through judicial review, but most analysts suggest this trend has been exaggerated. By and large, the executive rules unchecked and unchallenged whoever is in power. This is very unhealthy state of affairs, which has been exacerbated in recent years by large parliamentary majorities and/or long periods of one party rule.

Abolish the Civil Service

So what is to be done? Well, when I begin by saying that part of me wanted to entitle this article ‘Sir Humphrey and the Swedish model’ you may have some idea of what is coming next. We should effectively abolish the existing civil service and replace it with two new groupings: a national public service and a government policy service.

I have only slowly come around to this position, mainly from observing that every preceding reform effort has failed. The Whitehall village of today is not that different in culture and behaviour from when it was first termed that some 30 years ago. Some of the language has changed, new structures and systems introduced, but very little has really changed.

The diagnosis of the ‘Next Steps’ report in 1988 was that various efforts at systemic reforms (the Rayner scrutinies and the Financial Management Initiative) had largely failed. The main reason was that the culture hadn’t changed. In order to change culture, the report implied, what was needed was a big structural shock—the creation of executive agencies. This was tried, but Whitehall even managed to subvert this radical step. While it led to some improvements in some services, it had little impact on the core.

The core task for a ‘civil service’ is, and has always been, to help ministers of a democratically appointed executive shape policy and its implementation. There is no clear reason why this has to involve the direct management of any particular services or indeed of any services at all. And I think there are good reasons why such direct management of some services actually gets in the way of the good management of all services.

So part of what I would advocate is the ‘Swedish model’: relatively small ministries with large service delivery bodies clearly separated from them and with their own ‘constitutional personality’. This would free up the civil service (my government policy service) to become much more expert at those policy things which it does not do well now—policy-entrepreneurship, at managing policy and delivery networks and about evaluating programmes and services.

Paradoxically, it is probably only by separating the Whitehall village from direct responsibility for service delivery that it might become effective at managing it. Managing complex networks and modern ‘governance’ is a very different skill set from managing hierarchies. But the relatively small responsibilities Whitehall currently has for direct management of services tends to infect the way they view all other services. This is one important source of the much complained about ‘command-and-control’ culture. Anyone who has been around Whitehall knows that the language of ‘brigading’ staff, for example, is still endemic. Taking away direct services just might provide a big enough shock to the system to jolt it out of this ingrained behaviour.

There is an obvious inherent danger in such a strategy—separation could easily become isolation. But Whitehall is already extremely isolated from 90% of service delivery.

When it talks about ‘opening up’ it always means opening up to the private sector: hence the Whitehall and Industry Group. The UK has one of the lowest exchange rates between central government and other tiers of government and wider public services than just about any country I can think of. The Whitehall village is already cut off from the massive conurbation that is public services in Britain today. Paradoxically, again, it is only by completely separating that the village might just be forced to start engaging on more equal and realistic terms.

Specific measures would be needed to ensure that Whitehall does engage more effectively—including a substantial inter-change programme between Whitehall and wider public services. Stipulations that no-one can reach the top three tiers without at least five years’ experience in the public service outside Whitehall would be sensible.

Although this would be a ‘government policy service’ it should still be subject to greater checks and balances than currently exist. There is a strong case for at least partially opening up that supposedly sacrosanct arena of ‘policy advice to ministers’ to greater transparency and accountability through parliament. New Zealand, so celebrated for its managerial reforms, has taken steps in this direction without collapsing into anarchy.

So what becomes of the existing civil service agencies delivering services? I would propose the creation of a new category of organizations to be called a ‘national public service’ to include all existing executive agencies, NDPBs, public corporations and the NHS. (I leave aside the issue of local government, but this should be considered.)

All such bodies should be put on a statutory basis or at the very least in some way authorized by, and periodically re-authorized, by parliament. The latter should take a much more active part in scrutinizing their budgets and performance, along the lines of the Government Performance and Results Act 1993 (GPRA) in the USA. And the national public servants should be able to be held to account in parliament, in a way that civil servants are not currently constitutionally able to be.

Reforms to audit, inspection and regulatory bodies follow. The restrictions on the NAO’s ability to question ‘policy’ should be re-examined. The distinction between ‘policy’ and ‘value for money’ is almost impossible to establish and it makes the NAO overly cautious on some issues. The proposal to make the NAO responsible for validating performance information should also be enacted. Indeed, some have gone further and suggested a comprehensive performance review (CPA)-type system as applied in local government could be applied to national government too. There would certainly be much less reason to object to this for the national public service.

I want to conclude with a small story which, I think, illustrates almost everything that is wrong with the current civil service. A few years ago I was at a presentation by a very senior civil servant. In the question-and answer session I challenged the publication of the government’s ‘annual report’. I pointed out it was selective, un-audited and was undermining the government’s credibility rather than improving it.

The mandarin approached me afterwards and informed me quietly that the annual report problem was being dealt with—they were going to stop publishing it! There spoke a true serial monogamist, rather than a public servant—more concerned with protecting the government of the day than with transparent and accountable public service. I do not   blame individuals but the system which allows and encourages this to happen. It is time the civil service became a true public service.

Colin Talbot is at the Centre for Public Policy and Management, Manchester Business School.