. 110
( 155 .)


Parliament for policy, and any extension of the accountability of civil servants must recog-
nise the overriding responsibility of the Departmental Minister for the work and efficiency
of his department. The Government do not therefore favour developments which would
detract from the principle that the advice tendered to Ministers by civil servants should be
confidential and objective. . . .

The traditional view of the constitutional position of civil servants was
restated in 1985, in a Note by Sir Robert Armstrong, Head of the Home Civil
Service, issued after consultation with the Permanent Secretaries of government
departments. The Note, as revised in 1987 (HC Deb vol 123, cols 572“5w,
2 December 1987), declared:

The Civil Service as such has no constitutional personality or responsibility separate from the
duly constituted government of the day. It is there to provide the Government of the day
with advice on the formulation of the policies of the Government, to assist in carrying out
the decisions of the Government, and to manage and deliver the services for which the
Government is responsible. . . . In the determination of policy the civil servant has no con-
stitutional responsibility or role distinct from that of the Minister.

Professor Vernon Bogdanor said of the Armstrong Code that it ˜failed . . . to
take account of the fact that the role of the civil servant was changing with the
establishment of agencies [above, p 409] and other developments requiring
o¬cials to be far more involved in policy initiatives than traditional doctrines
would allow™ (Memorandum to the Select Committee on the Public Service:
Special Report, HL 68 of 1996“97, p 36, para 10). Although the Armstrong
Memorandum has been superseded by the Civil Service Code (above, p 419),
it has been declared by the Cabinet O¬ce to be still ˜a valuable statement of
constitutional principles™ (Public Service Committee, Second Report, HC 313-I of
1995“96, p xiii, note 13). Governments have continued to insist that civil servants
owe their loyalty to the duly constituted Government and are accountable to the
minister in charge of their department and not to Parliament. ˜The way in which
our constitution works™, said Sir Robin Butler to the Public Administration
Committee in 1997, ˜is that the Minister accounts to Parliament and the civil ser-
vants account to the Minister™ (Minutes of Evidence, HC 285 of 1997“98, Q70).
The absence of direct constitutional responsibility of civil servants to
Parliament is, however, not total and has been quali¬ed by recent changes in the
machinery of government.
586 British Government and the Constitution

A direct personal responsibility of o¬cials to Parliament has for many years
been formally acknowledged in one instance: the Accounting O¬cer of a
department, who is usually the Permanent Secretary, is responsible for the
departmental accounts and answers to the Public Accounts Committee of the
House of Commons for the regularity and propriety of departmental expendi-
ture and the observance of proper economy. This responsibility quali¬es the
Accounting O¬cer™s duty to the minister. If the Accounting O¬cer believes that
any projected departmental expenditure would be irregular or improper, it is
his or her duty to make a formal objection to it. The minister may override the
objection by giving the o¬cer a written direction, but the matter is then
reported to the Comptroller and Auditor General and the Treasury. A similar
procedure applies if the Accounting O¬cer is concerned that proposed action
may not achieve value for money. (See the Ministerial Code (2005), paras
3.2“3.3.) In 1991 a departmental Accounting O¬cer submitted a memoran-
dum of dissent to the Minister for Overseas Development, objecting that expen-
diture on the Pergau hydro-electric scheme in Malaysia would not be a prudent
and economic use of aid funds. He was overruled by the minister. Afterwards
the Accounting O¬cer appeared before the Public Accounts Committee to give
a full account of the circumstances: Public Accounts Committee, Seventeenth
Report, HC 155 of 1993“94. (For the subsequent judicial review of the minister™s
decision see below, p 662.)
Senior civil servants are frequently called upon to appear before other
parliamentary select committees, in particular the ˜departmentally related™
committees, and may be questioned about the work of their respective depart-
ments. Their evidence is constrained by ministerial responsibility and is subject
to limits established by the government, as we shall see, but the Ministerial Code
(2005) says that ministers should ˜require civil servants who give evidence
before Parliamentary Committees on their behalf and under their direction
to be as helpful as possible in providing accurate, truthful and full information™
to the committees. A considerable amount of information about the detail
of departmental administration is provided to the committees by o¬cials.
˜In practice™, as Peter Barberis observes, ˜civil servants do answer to Parliament,
most visibly through select committees™, and he adds that the load of explana-
tory accountability ˜is now borne by civil servants as well as by ministers™
(The Civil Service in an Era of Change (1997), p 144). Government, however,
withholds formal recognition from this development: ˜The Government™s
commitment to a permanent, non-political civil service means that there can
be no question of apportioning between the Minister and his civil servants
part or parallel shares in a single line of accountability to Parliament™ (Public
Service Committee, First Special Report (Government Response), HC 67 of
1996“97, p vi).
A direct responsibility of civil servants has from time to time been exacted
by committees or tribunals of inquiry which have identi¬ed civil servants as
being to blame for administrative failures. The Scott Report on arms to Iraq
587 Parliament and the responsibility of government

(HC 115 of 1995“96) made numerous criticisms of individual civil servants for
errors of judgement, neglect of duty, want of frankness and other ˜thoroughly
reprehensible™ conduct. (Disciplinary action was subsequently taken against
two o¬cials in the Foreign and Commonwealth O¬ce: Public Service
Committee, Minutes of Evidence, HC 285 of 1997“98, Appendix 2.) In 1999 the
Foreign A¬airs Committee in its report on the ˜Sandline™ a¬air (above, p 584:
Second Report, HC 116-I of 1998“99) concluded that certain named o¬cials
had ˜failed in their duty™ in not keeping ministers informed of events “ a judge-
ment not accepted by the Government (see its Response to the Committee, Cm
4325/1999). Subsequently the Inquiry into the Conservative Government™s
handling of the BSE epidemic found that there had been ˜institutional and
political failure up to the highest levels™, naming and criticising both senior
civil servants and ministers (Report of the BSE Inquiry, HC 887-I of 1999“00).
The conduct of o¬cials came under scrutiny in both the Hutton and Butler
reports in 2004 consequent upon the invasion of Iraq (HC 247 and HC 898 of
2003“04). No individual o¬cial was found to have been seriously culpable: in
particular, the misleading dossier of September 2002 on Iraqi weapons of mass
destruction, drawn up by the Joint Intelligence Committee under the chair-
manship of John Scarlett, was said by Butler to be the result of collective fail-
ures, such that Scarlett should not be expected to resign as head of the Secret
Intelligence Service.
The establishment of executive agencies under the ˜Next Steps™ programme
(above, p 409) has given civil service managers or ˜chief executives™ of the agen-
cies an enhanced responsibility and independence in carrying out their tasks.
The Treasury and Civil Service Committee perceived the implications of this
development for democratic control and accountability, saying in its Eighth
Report for 1987“88 (HC 494-I, paras 46“7):

The traditional system of accountability does not seem to us to be entirely consistent with
the increased delegation of responsibility to individual civil servants. . . . [T]hose who are to
make the decisions should be publicly answerable for them. . . . We certainly do not advo-
cate abandoning the principle of ministerial accountability, but modifying it so that the Chief
Executive who has actually taken the decisions can explain them, in the first instance. In the
last resort the Minister will bear the responsibility if things go badly wrong and Parliament
will expect him or her to put things right, but the process of Parliamentary accountability
should allow issues to be settled at lower levels, wherever possible.

The Government agreed to the Committee™s recommendation that chief exec-
utives of agencies should be appointed as Accounting O¬cers, with direct,
personal responsibility to the Public Accounts Committee for the use of public
money (see above) and that they should appear and give evidence before
other parliamentary select committees (albeit ˜on behalf of ministers™) on the
matters delegated to them (Civil Service Management Reform: The Next Steps,
Cm 524/1988, p 9).
588 British Government and the Constitution

Executive agencies operate under published framework documents des-
cribing their organisation and responsibilities and they publish annual reports
and accounts as well as three-year business plans which set out their objectives,
related to three-year funding agreements with their departments. Chief
executives are set publicly announced performance targets and are responsi-
ble to ministers for achieving them and generally for the management of the
agencies. Ministers remain responsible for the broad issues of policy relating
to agencies, the targets set for them and the resources provided.
Correspondence from MPs and written parliamentary questions which bear
on the functions delegated to agencies are referred by ministers to the chief
executives for reply. (Their replies to referred parliamentary questions are
published in the O¬cial Report (Hansard).) An MP who is dissatis¬ed with
a chief executive™s response can raise the matter with the minister, as being
ultimately accountable.
These extensive delegations to agency chief executives have intensi¬ed debate
about the location of responsibility to Parliament. Governments remain ¬rm in
their insistence on the traditional doctrine and have refused to accept argu-
ments that chief executives should be personally and directly accountable to
Parliament for the matters assigned to them. (See Taking Forward Continuity
and Change, Cm 2748/1995, p 31; Government Response to House of Lords Select
Committee on the Public Service, Cm 4000/1998, paras 14, 44“6.) Perhaps Lord
Mackay of Ardbrecknish, a minister of state, was not mindful of the govern-
ment™s o¬cial standpoint in saying to the Public Service Committee of the
House of Lords: ˜We [ministers] are accountable to Parliament. And the Chief
Executives themselves, perhaps at a non-policy level, at the administration level,
are also accountable to Parliament™ (Public Service Committee, Special Report,
HL 68 of 1996“97, Ev, Q 885). O¬cial doctrine and reality seem to diverge in
the evolving relations of ministers and their agencies to Parliament.
The division of responsibilities between ministers and their agencies is
expressed in o¬cial pronouncements in terms that responsibility for policy
is owed by the minister to Parliament, while operational responsibility is dele-
gated to the chief executive and is owed to the minister. But the distinction
between ˜policy™ and ˜operations™ is problematic and becomes blurred in prac-
tice, since ministers have authority to intervene in operational matters and
have sometimes done so frequently and in detail. The formula does nothing to
illuminate the shadowy contours of ministerial responsibility. Christopher
Foster remarks that the di¬usion of power in government, of which executive
agencies are an instance, ˜has di¬used responsibility and made it harder to
apportion blame fairly™ (British Government in Crisis (2005), p 153). A report
by the Institute for Public Policy Research (Whitehall™s Black Box: Accountability
and Performance in the Senior Civil Service (2006)) argues for a reformulation
of the convention of ministerial responsibility, to be incorporated in a Civil
Service Act, which would provide for a clearer division of responsibilities
between ministers and civil servants. The accountability of civil servants would
589 Parliament and the responsibility of government

be assured by new governing arrangements for the civil service, the permanent
secretaries of departments being held personally accountable for departmental
operations to a new Head of the Civil Service.
The actions of civil servants may be the subject of investigation by the
Parliamentary Commissioner for Administration (Ombudsman) (see further
below). The Ombudsman™s inquiries into maladministration by government
departments may lead to ¬ndings that particular civil servants have been at
fault, but those concerned are not usually named in the resulting report to
(See generally D Woodhouse, Ministers and Parliament: Accountability in
Theory and Practice (1994); P Giddings (ed), Parliamentary Accountability: A
Study of Parliament and Executive Agencies (1995); A Tomkins, The Constitution
After Scott (1998); Barberis, ˜The new public management and the new account-
ability™ (1998) 76 Pub Adm 451; P Riddell, Parliament under Blair (2000), ch 4;
Hogwood, Judge and McVicar, ˜Agencies and accountability™, in RAW Rhodes
(ed), Transforming British Government (2000); Elder and Page, ˜Accountability
and control in next steps agencies™, in Rhodes, ibid; Hansard Society
Commission, The Challenge for Parliament: Making Government Accountable
(2001); Woodhouse, ˜Ministerial responsibility™, in V Bogdanor (ed), The
British Constitution in the Twentieth Century (2003).

3 The power of Parliament
We are not now concerned with the ˜sovereignty™ of Parliament, or that
supreme law-making power which belongs not to Parliament alone but to the
Queen in Parliament, and which in reality is mainly at the disposal of the
government. Our present interest is in the power of Parliament, in particular
the House of Commons, to perform its functions of controlling and scrutin-
ising the executive “ these terms being taken in a wide sense: ˜controlling™ to
include in¬‚uencing or restraining, and ˜scrutinising™ (or calling to account)
to include extracting information, criticising and procuring reparation or
redress. In carrying out these tasks, Parliament relies less on its formal powers
(eg to enforce the production of papers or punish for contempt) than on the
conventional responsibility owed to it by ministers, and the practices and
procedures that have crystallised about this convention.

Bernard Crick, The Reform of Parliament (rev 2nd edn 1970),
pp 79“81

Politics, not law, must explain the concept and practice of Parliamentary control of the
Executive. In modern conditions any such control can only be something that does not
threaten the day-to-day political control of Parliament by the Executive. The hope for any
worth-while function of control by Parliament would be grim indeed if it depended on the
590 British Government and the Constitution

ultimate deterrent of the vote: the undoubted Constitutional right of Parliament to vote
against the Queen™s Ministers and the Convention by which they would then resign. But
control, on both sides, is indeed political. Governments respond to proceedings in Parliament
if the publicity given to them is likely to affect public confidence in the Government, or even
if the weakness with which the Government puts up its case, even in purely Parliamentary
terms, begins to affect the morale of its own supporters (though it takes a very long suc-
cession of bleak days for the Government in the House before the country begins to be
The only meanings of Parliamentary control worth considering, and worth the House
spending much of its time on, are those which do not threaten the Parliamentary defeat of
a government, but which help to keep it responsive to the underlying currents and the more
important drifts of public opinion. All others are purely antiquarian shufflings. It is wholly
legitimate for any modern government to do what it needs to guard against Parliamentary
defeat; but it is not legitimate for it to hinder Parliament, particularly the Opposition, from
reaching the public ear as effectively as it can. Governments must govern in the expectation
that they can serve out their statutory period of office, that they can plan “ if they choose “
at least that far ahead, but that everything they do may be exposed to the light of day and
that everything they say may be challenged in circumstances designed to make criticism as
authoritative, informed and as public as possible.
Thus the phrase ˜Parliamentary control™, and talk about the ˜decline of Parliamentary
control™, should not mislead anyone into asking for a situation in which governments can
have their legislation changed or defeated, or their life terminated (except in the most des-
perate emergency when normal politics will in any case break down, as in Chamberlain™s
˜defeat™ in 1940). Control means influence, not direct power; advice, not command; criticism,
not obstruction; scrutiny, not initiation; and publicity, not secrecy. Here is a very realistic
sense of Parliamentary control which does affect any government. The Government will
make decisions, whether by existing powers or by bringing in new legislation, in the
knowledge that these decisions, sooner or later, will find their way to debate on the Floor
of one of the Houses of Parliament. The type of scrutiny they will get will obviously affect,
in purely political terms, the type of actions undertaken. And the civil service will adminis-
ter with the knowledge that it too may be called upon to justify perhaps even the most
minute actions . . .
Governments deserve praise in so far as they expose themselves, willingly and helpfully,
to influence, advice, criticism, scrutiny, and publicity; and they deserve blame in so far as
they try to hide from unpleasant discussions and to keep their reasons and actions secret.
Parliaments deserve praise or blame as to whether or not they can develop institutions
whose control is powerful in terms of general elections and not of governmental instability.
This ˜praise™ and ˜blame™ is not moralistic: it is prudential . . . So Parliamentary control is not
the stop switch, it is the tuning, the tone and the amplifier of a system of communication
which tells governments what the electorate want (rightly or wrongly) and what they will
stand for (rightly or wrongly); and tells the electorate what is possible within the resources
available (however much opinions will vary on what is possible) and “ on occasion “ what is


. 110
( 155 .)