Written evidence from Professor David Richards and Professor Martin J. Smith[i] (WCS 03)

Public Administration and Constitutional Affairs Committee

The work of the Civil Service inquiry

• This evidence below while specifically addressing the fourth question concerning the issue of Civil Service impartiality and the sustainability of the Haldane doctrine, nevertheless reflects on a number of issues of concern to this Inquiry
• We draw on a range of evidence to highlight how Whitehall reform over the last four decades but most notably the post-2010 period, has led to the abandonment in all but name of the Haldane doctrine.
• In its place has emerged a more universal command and control [re. principle-agent] relationship between ministers and civil servants which is seen as necessary to meet the demands of twenty-first century bureaucracy.
• This in turn creates a crucial problem; whilst UK governance is going through a process of radical change, the constitutional precepts are said to remain in place.
• The net effect of the political class holding to this position is that the mechanisms of democracy and accountability are not keeping up with changes in practice.
•                    We conclude by reflecting on the pathologies that have emerged from the inconsistency of trying to sustain such a position.

Overview: The Haldane Model - Then and Now

1. The central convention underpinning the Westminster model [WM] which operated throughout much of the twenty-century is the notion that the relationship between a minister and civil servant is symbiotic and indivisible. 
2. First set-out in Lord Haldane’s 1918 Report, it established the convention that officials advise a minister on a subject [free from fear or favour], so there was no requirement for the separation of power between the political and administrative class.
3. From this principle flowed the constitutional convention that ministers and their officials had a ‘shared personality’ and were seen as a single entity.  A convention that received legal clarification in the 1944 Carltona doctrine (Richards 2008).
4. In practice, what evolved was a model of ministerial power expressed through the department with officials implementing a minister’s wishes.  The norms underpinning the senior civil service included: political neutrality; dedication to serving both the government and the national interest [which at times creates tension]; loyalty to the minister but being afforded the deliberative space to [critically] advise; to serve all governments equally; and accountability to the minister who in turn was accountable to Parliament - officials had no public personality, they were anonymous. It is on these conventions that the WM was established over a century ago.
5. The last four decades have witnessed an almost continual programme of Whitehall reform, often discussed under the label of New Public Management (NPM) which has cumulatively led to the diminution of the traditional Haldane understanding of the minister-civil servant relationship.
6. What has emerged in its place is a more universal command and control, principle-agent relationship which is seen as necessary to meet the demands of modern accountability and transparency.
7. The pace of this transformation has notably accelerated since 2010 and this in turn has revealed a key issue; whilst UK governance is going through a process of radical change, the constitutional precepts are said to remain in place.
8. The net effect of the political class holding to this position is that the mechanisms of democracy and accountability are not keeping up with changes in practice.
9. A more specific pathology exposed by the shift in the nature of the relationship between ministers and civil servants to a principle-agent setting is that the deliberative space afforded for critical engagement over public policy has been diminished.
10. The WM has always been recognised for its limited number of veto points. Such a shift has further exacerbated this dimension of the model, leading to what has been referred to elsewhere as a growth in the “blunders of our governments” (King and Crewe 2013).
11. A critical gap now exists between the actual practices of government and the constitutional architecture – particularly over the issue of both accountability and checks-and-balances, in terms of the senior civil service as a veto-player

Challenges to Haldane 1979-2010

12. From 1979-97, governments politicised the ministerial and civil service relationship, not through a process of “overt” party politicisation (see Richards 1997) but covertly in the way in which a “them” and “us” relationship emerged.
13. The Thatcher and later Major Governments implicitly rejected the Haldane model, seeing officials not as facilitators, but as a constraint on the achievement of ministerial goals. This in turn led to a series of reforms concerned with bolstering ministerial power and a concomitant erosion of Whitehall’s veto-playing capabilities.
14. Reform, informed by public-choice accounts of bureaucracy, depicted the Civil Service in a more adversarial light. The idea that officials had a role in mediating ministerial preferences was challenged head-on. This period saw a shift in the minister-civil servant relationship from one of mutual dependence to that of a principle-agent mode. 
15. It should be recognised that a principal-agent model generally assumes there are conflicts of interest between the principal and the agent.  It is therefore a model that sets itself apart from the Haldane convention.
16. From 1997-2010, Whitehall reform continued unabated and led to some notable changes:

• significant increases in special advisers (especially around the prime minister);
• a much greater focus on implementation through the Prime Minister’s Delivery Unit;
• increased use of private suppliers in the delivery of public services
• greater use of targets and market mechanisms as means of controlling public servants, both in and beyond Whitehall.

17. As evidence to previous inquires reveal, reform led to a growing mistrust between officials and ministers (although this was not universally the case) and with it a further strain on the Haldane convention.  Yet change was presented as occurring within the context of the existing constitutional arrangements.

Challenges to Haldane 2010-onwards

18. Since 2010, governments have continued to try and have it both ways – pursuing far reaching reform while claiming the constitutional conventions underpinning Whitehall remain unaffected. 
19. The pathology of such an approach is in the ambiguity and confusion it creates. The last Coalition Government’s civil service reform plan neatly captures such inconsistency, seeing no irony in suggesting that everything will stay the same, yet everything must change:

The current model of a permanent, politically impartial Civil Service will remain unchanged. It exists to serve the Government of the day, while retaining the flexibility to serve future Governments. Civil servants carry out three broad types of role, at home and overseas. All need to change in order to meet the new challenges and address weaknesses. [emphasis added] (HM Government 2012, 8).

20. The brief survey of key reforms below highlights how far reaching the recent reform process has been and in so doing reveals the emergence of an accountability-based, principal-agent relationship between ministers and civil servants that stretches beyond credulity those that suggest  Haldane remains intact.
21. The rise in political advisors – 38 in 1997, 74 in 2010 and 107 by 2015  and the agreement to create of extended ministerial offices [ EMOs] have increased ministerial resources in relation to Whitehall.
22. There has also been an increased role for ministers in appointing Permanent Secretaries. This change raises questions over the principle of neutrality and whether senior officials seeking appointment at this level would continue to “speak truth unto power” for fear of promotional non-preferment. 
23. The introduction of performance objectives for individual Permanent Secretaries in 2012 challenges the WM’s conception of individual ministerial accountability. 
24. The introduction of Senior Responsible Owners (SROs) for major projects now make SRO civil servants directly accountable and changes the Osmotherly rules.  Whether this reform works well or goes badly (as in the past with attempts to make Next Steps Agency Chief work to a framework document), either way it will draw attention to the myth of ministerial responsibly.
25. The effect has been a recalibration in the relationship between ministers and civil servants in which ministers have sought to increase their power over the senior civil service through particular mechanisms of managerialism and accountability. 
26. Such reforms are not without contradiction. On one level, ministers have berated Whitehall for its lack of skills and blamed civil servants for blocking change (a view consistent across all governments since 1997). On another level, ministers see reform occurring through officials taking more responsibility – the desire for “delegated mission command” (PASC 2013).
27. One consequence of the pursuit of managerialist reforms across Whitehall has been to reduce the policy advice function of officials. Civil servants require different skills for the 21st century, particularly in policy implementation [see Table 1].

Table 1. Changing Skills of Civil Servants

28.                   Civil servants are still trained as policy makers, but increasingly they work in management, contract negotiation, project management and implementation.
29.                   Two features emerge from this arrangement.  The first is de-emphasising the critical, deliberative function of senior officials in policy-making, manifested in a principal-agent approach. The second is a skills gap, illustrated by a number of high profile policy failures (see King and Crewe 2013). Officials are identified as lacking the ability to oversee complex contract negotiations. The approach to large and complex problem is not defined as an issue of “project management”, as it would be in the private sector, but instead is seen in terms of delivering the normal run of business (National Audit Office 2014/15).
30.                   The Coalition sought a distinct break from the target culture associated with the New Labour, yet it has re-emerged in all but name.
31.                   Public Service Agreements and Departmental Strategic Objectives have been replaced by Departmental Business Plans offering a different form of control and accountability, eschewing the language of targets. Priorities and transparency became the new lingua franca of public services.
32.                   Government and public sector web sites were identified as the medium to deliver a new form of democratic accountability. Yet targets in all but name have remerged (see Table 2).

Table 2: Numbers of Departmental Targets/Objectives 2010, 2011-12 and 2015

33.                   Departmental business plans operating in conjunction with Permanent Secretary objectives symbolise the extent to which an informal culture of targets based on a principal-agent model remain firmly embedded in Whitehall.
34.                   Rhetorical claims that bureaucratic accountability has been abandoned at the behest of democratic accountability appear overblown.              
35.                   Thirty years of managerialist reforms which have accelerated since 2010 have not led to the abandonment of centralism, even in an age of devolution. Ministers have sought to further assert their authority over Whitehall and large swathes of the extended state by employing a principal-agent approach in the delivery of public goods.
36.                   To explain this recurrent contradiction the focus lies with ministerial responsibility and the way it shapes politics and policy. The Westminster convention that ministers have responsibility for all that goes on in their domain invokes a strong reflexive tendency in the British polity against any drift towards subsidiarity.
37.                   The adversarial nature of the British political system combined with ministers being responsible for all they survey, ensures a resistance to download key policy decisions to localities or managers. This is a consequence of the way ministers sit within their departments.
38.                   Civil servants are there to protect and support ministers (an example of bureaucratic accountability) not to serve the public (democratic accountability). Officials are adept at developing policy set by ministers, defend ministerial positions and generally supporting departmental lines and budgets. They are less well equipped to think about the development and delivery of policy on the ground.
39.                   This creates a frustration for ministers who find themselves drawn to the support officials provide, but are disappointed when policies are not effectively implemented.
40.                   Frustration over issues of both implementation and delivery, the so-called ‘rubber lever syndrome’, has led to ministers seeking to hold Whitehall to account for such failings. The use of targets (“objectives”) for both departments and permanent secretaries is the most obvious manifestation of their response. It has led to a fundamental change in the relationship between ministers and civil servants; a shift from mutual dependence to a binary mode of separation, based on a principal-agent model.
41.                   Ministers have sought to download accountability while at the same time protecting against a diminution in the power they wield.
42.                   During this time, governments have avoided addressing questions on the constitutional implications implied by these reforms. Their approach has been to argue that reform is wholesale, yet maintaining that the WM remains intact. Anything else would require the need to re-imagine the long-established constitutional certainties underpinning the British approach to governance.

Whitehall’s Future and the Pathology of Sustaining the Myth of the Westminster Model

43.                   This final section identifies a set of unresolved tensions across Whitehall that in the longer-term appear unsustainable.
44. A number of pathologies emerge from a strategy emphasising the fundamental reform of Whitehall while claiming constitutional conventions are unaffected.
45. The structures of government based on functional departments and processes of accountability through ministers and departments remain largely nineteenth century entities. Consequently, there is a growing dissonance between the way policy is made and delivered and the institutional and democratic structures of government.
46. There is a fundamental contradiction in relation to expectations of implementation. Government is expecting, through targets and benchmarks, consistency in national standards, while encouraging diversity in delivery at the local level (education policy is an obvious example where government has created free schools and academies but then used inspection and sanctions on head teachers to ensure national standards). This is where the preservation of the WM is revealed: managers are free to manage as long as they do what central government requires. No government has yet reconciled the desire for decentralised implementation beyond Whitehall with the desire to retain ministerial control.
47. The problem of accountability is far from being resolved. The WM offers an appealingly simple, lineal, but outdated model of accountability. There has been no attempt to construct a process of accountability when autonomy exists at different levels of policy making and delivery.
48. How, for instance, do you hold government to account when contracts outlast the life of an administration? Under the new forms of governance there is a need for contractors, voluntary sector organisations, officials and ministers to define their responsibilities and then to have clear lines of accountability that relate to the reality of government rather than a constitutional myth.
49. The vision of the Civil Service operating in what the current Government refers to as a ‘post-bureaucratic’ world is very different to the nineteenth century Northcote-Trevelyan paradigm. In an environment of pluralised policy-making and delivery, officials need to become the arbiters of the system ensuring transparency, fairness and accountability in a fractured system of governance.
50. Officials need new skills, not just as project managers, but as contract specialists.  As one example from many, the upgrading of London Underground, officials lacked the skills to negotiate contracts and the situation was made worse by the limited number of major accountancy firms overseeing the process who often provided advice both to contractors and government.
51. The role of the Civil Service is potentially shifting towards that of an ‘overseer’. It needs to oversee contracts, accountability and democratic processes, and ensure equity in a devolved and fragmented system. The primary function of officials is no longer just that of policy making, but now also as regulators of the policy process.  Given the absence of veto-players in the Westminster system, this becomes a crucial dimension.
52. Yet, the evidence presented here and elsewhere suggests that reforms and their impact on the minister-civil servant relationship is constraining the latter’s veto-playing function.  A National Audit Office (2016: 4) report on accountability concludes:

The complexity of both policy and implementation has increased over the decades, the balance of pressures on AOs (Accounting Officers) has shifted in a way that potentially undermines accountability to Parliament. AOs now operate in an environment where ministers often perform a more ‘executive’ role in policy implementation and have sought greater involvement in top civil service appointments, while appointing increasingly influential special advisers to act on their behalf. This appears to have tilted the balance so that AOs have greater pressures to give weight to political drivers rather than public value. [italics added]

53. The Report led the former Chair of the Public Administration Select Committee, Margaret Hodge (2016) to conclude that: ‘…the old convention of civil servants being accountable to ministers who are accountable to parliament is broken’. It is a view we concur with supported by the evidence we have offered above. 

54. To address these issues, change could for example involve an organisational and cultural shift from a departmental-focused to a problem-focused approach which is more flexible, smaller and digitally based.

55. This would require a move from hierarchical bureaucracy to what we would label “cybernetic-squad” bureaucracy. A recognition that in a digital age, there are certain, specialised bureaucratic/technical skills that are not linked to departmental functionalism, but need to focus on specific problems, allow for flexibility and rapid intervention.

56. Expertise would no longer be located within a department, but to a team deployed to deal with particular projects or intractable problems. This was how the now defunct Implementation Unit operated - identifying problems and seeking to resolve them. Such an approach would allow a concentration of the new skills required of civil servants and avoid duplication.

57. It would, however, require a very different ministerial-civil service relationship and more particularly, an overhaul of the Westminster Model.

References

HM Government. 2012. The Civil Service Reform Plan June 2012. London: The Stationary Office.

Hodge, Margaret 2016. ‘Accountability system for civil servants is “broken” — but MPs are more interested in new policies than value for money’ Civil Service World 4 MRCH 2016 Available online at: https://www.civilserviceworld.com/articles/news/margaret-hodge-accountability-system-civil-servants-“broken”-—-mps-are-more-interested

King, Anthony and Ivor Crewe. 2013. The Blunders of our Governments. London: Oneworld.

National Audit Office 2016. Accountability to Parliament for Taxpayers’ Money HC 849 Session 2015-16 23 February 2016 Available online at: https://www.nao.org.uk/wp-content/uploads/2016/02/Accountability-for-Taxpayers-money.pdf

Public Administration Select Committee. 2013. Truth to power: How Civil Service Reform can Succeed. Eighth Report of Session 2013–14, 6 September 2013. London: The Stationery Office.

Richards, David 1997. The Civil Service under the Conservative 1979-1997 Brighton: Sussex Academic Press

Richards, David. 2008. New Labour and the Civil Service: Reconstituting the Westminster Model. Basingstoke: Palgrave Macmillan.