In 1984 Prime Minister Bob Hawke was presented a challenge: how to maintain fundamental political conventions while solving a pressing political crisis. One of his cabinet ministers, Stewart West, objected to a cabinet decision on the mining and export of uranium. To protect his left-wing credentials he wanted to resign from cabinet but remain as a minister, so that he could oppose the decision in the Labour caucus. But Hawke had declared, on becoming prime minister, that all ministers would be bound by collective responsibility in all forums, including the caucus; that was no more than a standard interpretation of the convention. He asked the secretary of the Department of the Prime Minister and Cabinet, Geoff Yeend, for assistance.
Yeend worked with his colleagues to develop a definition of collective responsibility that allowed the minister to make his opposition known in caucus without his having to resign from the government. A revised version of the Cabinet Handbook (1983) blandly declared:
2.1 The convention of the collective responsibility of Ministers for Government decisions is central to the Cabinet system of government. Cabinet Minutes reflect collective conclusions and are binding on Cabinet Ministers as Government policy both outside the Party and within. This applies also to non-Cabinet Ministers co-opted to attend Cabinet meetings, in respect of matters dealt with while they are present.
2.2 Minutes of the Ministry are binding on all Ministers, similarly, both outside the Party and within, in respect of issues handled by the Ministry.
2.3 All Ministers are expected to give their support in public debate to decisions of the Government; non-Cabinet Ministers, however, are not prevented from debating in Caucus decisions in areas apart from their portfolios. Caucus decisions are binding on all Ministers.
The additional last interpretation in the clause thus released some ministers from collective responsibility within the party. A minister had to be present in cabinet before he was bound by the decision on caucus. In effect the change allowed West to resign from cabinet, but not from the ministry! Such an interpretation undermined traditional understandings of the convention, but no one cared. It was a device that accommodated internal difficulties, and then required neat drafting to bring the description of the convention in line with the practice.
The incident illustrates two fundamental facets of cabinet government. First, the great traditional conventions: individual ministerial and collective responsibilities exist, but they are flexible and open to interpretation. Their interpretation is contingent on circumstance and prime ministerial interpretation. Second, the rules of cabinet, whether set out in cabinet handbooks or manuals, are in a sense a formulation of the existing practices; they spell out how the traditional conventions will be applied for the time being, and they are as often designed to reflect as much as to lead practice and are vehicles through which prime ministers adjust their management of government. In this instance, the clause was set out as though nothing had changed, as though the Handbook were merely stating the obvious, accepted and long-standing practice. But of course it was not; it allowed a flexibility that had not previously existed for the Hawke government, but which had existed in the Whitlam government where ministers were allowed to re-fight in caucus battles they had lost in cabinet.
The new version of the Handbook points to a fundamental element of cabinet government. Cabinet government is based on convention, not law. Its significance and longevity is a tribute not to its formal power but to its ability to react to circumstance and to adjust procedures and practices to meet changing demands.
The conventional nature of cabinet should be no surprise, given its origins. It emerged slowly as the monarch ceased to turn up to meetings of his key council and as ministers became more answerable to the parliament than to the King. They agreed to agree in order to prevent the King from pursuing a policy of divide and rule (Mackintosh Reference Mackintosh1962). Gradually the monarch’s executive council became, in Bagehot’s (Reference Bagehot and Crossman1963) terms, a formal and dignified, but not a working, part of the constitution. Cabinet evolved and gathered authority precisely because there were no rigid rules that could be readily bypassed or gamed. It should be no surprise that it continues to evolve.
Indeed when the Australian Constitution was written, cabinet was noticeable for its absence. The colonial governments all had systems of cabinet government that they took as a given when they devised the constitution, but they preferred to describe the formal ‘dignified’ structures of monarch/governor and executive council, rather than try to define a way that cabinet should work. Prime ministers were absent from the constitution too, left to devise their own means of procedure.
This chapter deals primarily with the development of cabinet conventions in Australia and the United Kingdom. It could have readily looked at Canada and New Zealand because the conventions and practices have mutated there in similar ways, whether in the face of demands for transparency, for greater control or, more often, in reaction to political necessity. There are differences; they will be illustrated later by citing the other definitions of collective responsibility. The conclusions, however, would be similar.
Cabinet conventions
Cabinet conventions can be classified at two levels:
The broad constitutional conventions of ministerial and collective behaviour that notionally define the expectations of ministers in relation to the cabinet, the government and the parliament.
The rules and routines that in effect put those conventions into practice; they describe the way that cabinet is run, papers are considered and decisions made. These can be found in the cabinet handbooks, manuals, guidelines and codes of conduct that have been issued in the recent decades.
I argue that in neither case are the conventions fixed and uncontested. There is no absolute agreement on how the conventions should be applied in all cases. There is no set of rules for cabinet procedure that provide rigid structures that cannot be changed. Rather, at both levels of convention the prime ministers are the principal authorities that determine what the conventions mean, whether and how they should be applied, and when even normal procedures can be relaxed or ignored because it is convenient to do so.
In making decisions about the application of the conventions and rules, prime ministers must calculate how much can be altered. Ignoring and adjusting the conventions are not cost-free. How they apply the conventions and shape the rules, in what way they determine to enforce the rules of cabinet, when they decide to consult with which colleagues, all have important consequences for the running of government and the cordial relations with ministerial colleagues. The implementation of governing by rules is regarded as the foundation for good government.
The broad conventions
The core element of collective responsibility is easy to assert. Ministers may argue for or against a proposal in the privacy of the Cabinet Room, but once a decision has been made, they are required to support it in public, or at least acquiesce and say nothing. That convention is based on the assumption that they had an opportunity to put their case, that they lost the argument with their colleagues and that therefore they support the cabinet decision in public. The principle then goes further. Ministers must support cabinet decisions even if they were not present when the decision was made. Some cabinet committees are given the authority to make decisions on behalf of government; they have the same standing as those of cabinet. Ministers are equally bound to them, even if they were not involved in the debate and decision.
In all cases prime ministers can make exceptions, usually for solid political reasons. Hawke’s convenient reinterpretation of the doctrine has been noted. Harold Wilson relaxed the collective responsibility rules for the referendum on membership of the European Community. When the coalition government as formed in the United Kingdom in 2010 the Cabinet Office quietly released its new description of cabinet rules: ‘All government ministers are bound by the collective decisions of Cabinet and Cabinet Committees, save when it is explicitly set aside’ (United Kingdom Cabinet Office 2010).
So collective responsibility still applies, except where it does not! Here was another case of allowing political necessity and constitutional convention to find some practical common ground. Some concession had to be made to the coalition partners to make the cabinet process palatable. Prime Minister Cameron devised a cabinet system that took account of the expectations of his coalition partners and gave them an apparent share of influence. This was reflected in the constitution of the cabinet committees. Each cabinet committee has a chair from one party and a deputy chair from the other, and each has ‘the right to remit an issue to the Coalition Committee if it affects the operation of the coalition and cannot be resolved by the originating committee’ (United Kingdom Cabinet Office 2010). The four leaders, two from each party, made up the Coalition Committee. Even though the Coalition Committee was in form a cabinet committee, in practice it was a committee for politics. These arrangements were an indication of the pressures on normal practice created by coalition government.
Whether the actual processes worked the way they were originally planned is less important here than the insight they allow into the flexibility of cabinet conventions. It was after all a coalition; that was unusual in Britain and required novel forms of decision making to ensure that both partners felt they had a reasonable opportunity to put their case. The new arrangements were presented in terms that implied business as usual, as though, in terms of practice and constitutionality, nothing had changed. But they played havoc with what had been regarded as normal views of cabinet government and stressed yet again that collective responsibility is ultimately a flexible doctrine that can be adjusted to fit political circumstances, but in a way that still allows some meaning and underpins a coherent government. Leaders have to find a balance.
Ministerial responsibility too has its ambiguities. Ministers are answerable for the actions of their departments and their officials. If departments mess up, they need to explain and fix. They are not, never were and never should be required to resign only because a department has mismanaged a program. But once the problems are public and known, it is seen as reasonable that ministers should ensure that they are fixed. The oft-demanded idea that ministers should resign is no more than a distraction. Ministers will only contemplate resignation if their fingerprints are all over the area under attack, if they have given instructions that led to the circumstances, if they planned and oversaw its delivery.
If the fingerprints are the minimum condition, they do not by themselves constitute adequate cause. Resignation is normally a decision of the prime minister. Normally resignations are remembered for far longer than the cases where there was a crisis that did not lead to a departure. Prime ministers will calculate how long a crisis will continue and whether they want to keep the minister, as much as they will consider any intrinsic evidence.
Ministerial responsibility should be distinguished from ministerial probity. Many ministers are forced to resign because they misused travel accounts, gave advantage to friends and relatives, had conflicts of interest in share holdings that they forgot to mention, abused staff, lied to the prime minister or parliament or sundry other personal errors. We expect high standards of probity from ministers. Because they are clearly personally involved, there is little chance to slough off these errors or to blame someone else. Prime ministers again must use their judgement and decide whether they wish to tough it out. It may well depend on their opinion of the minister as much as the sin committed. These all-too-common cases are not constitutional issue of ministerial responsibility; they are just decent behaviour and standards. They still tend to get muddled.
The essence of these traditional constitutional conventions is their potential ambiguity or flexibility; their application is contingent on time, situation and the prime minister. When they seem to have been breached, the question is whether and what sanctions should be applied. The final judge and executioner will be the prime ministers. The prime ministers will receive plenty of advice, from academics, journalists and judges, and of course the opposition, about how and when the standards should be upheld.
But ambiguity does not mean these conventions are devoid of meaning or have no influence on ministers or cabinets. Practitioners are constantly aware of the potential impact of these grand constitutional conventions because they realise the political costs that are incurred by those who do not meet reasonable standards. Ministers who kept diaries and recorded the arguments in cabinet meetings often annoyed their colleagues who wanted to have a private debate and present a picture of public cohesion. That is stripped away by the records of division that diaries reveal. When Australian minister Neal Blewett, dedicated his Cabinet Diary (Reference Blewett1999): ‘to my cabinet colleagues in the hope that they will understand if not forgive’, he hit a chord. They were not pleased. There is a good political reason for a government to appear united.
The ministers are conscious too that they will be judged in public for their competence; as a consequence they take questions in parliament seriously too. They are aware that they can lose standing with their colleagues and backbenchers if they cannot perform well, whatever system of parliamentary questions is used. They relax when question time is over. Responsibility for their departments is not a myth but something real and pressing. So even if they can never explain with precision what will happen if these grand constitutional principles are abused, even if they do not know precisely what may be required in every circumstance, they are aware that the conventions are present, they matter and that the potential consequences are very real.
The conventions must remain ambiguous because it is impossible to write rules for all circumstances. Sometimes the interpretations will be very general indeed. Every country can have their own interpretation; it may well differ from decade to decade. I have considered those in Australia and the United Kingdom; it is worth a detour to look at other practices.
Canada is now the most direct. The rules are set out in Accountable Government: A Guide for Ministers and Ministers of State (Privy Council Office 2011). There is an introduction by Prime Minister Stephen Harper explaining the need for speaking with one voice. The rules themselves are blunt. There is no velvet glove, no doubt where power lies.
‘I.1 Individual Ministerial Responsibility
Ministers are accountable to the Prime Minister: they are appointed by the Governor General on the advice of the Prime Minister and the Prime Minister may ask for their resignation at any time.
Ministers are also accountable to Parliament…
I.2 Collective Ministerial Responsibility
All members of the Ministry are collectively responsible for carrying out the government’s policies as established by the Cabinet. They are therefore expected to work in close consultation with their ministerial colleagues. This principle is the foundation of a key constitutional convention known as Cabinet solidarity.
Policies presented to parliament and to the public must be the agreed policies of Cabinet. Ministers and Ministers of State cannot disassociate themselves from or repudiate the decisions of Cabinet or their Ministry colleagues unless they resign from the Ministry’
So, ministers, take care.
New Zealand’s Department of the Prime Minister and Cabinet (DPM&C) is gentler in its description, in part because many of its governments have been coalitions, and like the United Kingdom the practices have to take account of junior partners.
5.23 Acceptance of ministerial office means accepting collective responsibility. Issues are debated vigorously within the confidential setting of Cabinet meetings, although consensus is usually reached and votes are rarely taken. Once Cabinet makes a decision, Ministers must support it, regardless of their personal views and whether or not they were at the meeting concerned.
5.24 In a coalition government, Ministers are expected to show careful judgment when referring to party policy that differs from government policy. Subject to paragraphs 5.24–5.27 a Minister’s support and responsibility for the collective government position must always be clear
Paragraphs 24–26 talk about the agreement to disagree, but it is a permission given to coalition parties, not to individual ministers (DPM&C 2008: 65–6). Paragraph 5.27 explains that ministers in supporting parties outside cabinet are bound only in relation to their particular portfolios (DPM&C 2008: 66). There are therefore a number of gradations of collective responsibility, from cabinet ministers, to non-cabinet ministers from different parties, to ministers from supporting parties. No wonder the UK Cabinet Office looked to New Zealand for precedents when they had to devise rules of a coalition government. The value of the New Zealand formulation is illustrated by the fact that the cabinet manual has not been amended since 2008. Indeed the introduction is still under the signature of long-gone Prime Minister Helen Clark. The new government, it seems, saw no need to update them, or even re-issue them under the new prime minister’s name.
However, here is the most flexible interpretation I have ever found.
The Irish manual, Government Procedure Instructions (1983), provides a wonderfully elusive description of expectations:
18. The doctrine of collective responsibility requires that each Minister should inform his colleagues in Government of proposals he or a Minister of State at his Department intends to announce and, if necessary, seek their agreement. This applies, in particular, to proposals for legislation which can be initiated only after formal approval by Government. Ministers are responsible for implementing decisions.
And the Irish rules go on to explain ministerial responsibility:
23. It is the responsibility of the Minister concerned to ensure that Government decisions, whether formal or informal, are implemented at the earliest practicable date.
So then collective responsibility means ministers telling their colleagues what they propose to do, and ministerial responsibility means implementing government decisions. The concept of informal decisions, presumably what ministers chose to think they had been authorised to do, but without a formal record, would provide some ministers with great scope for initiative!
The rules of cabinet procedure
The challenge for leaders is how to make these conventions work. Over the recent decades a series of rulebooks have been developed, in part by bureaucracies, in part by prime ministers, to spell out the expectations of how ministers should act. These rulebooks are, in a real sense, the exposition of the constitutions in working form. Two examples of their gradual development can illustrate how rulebooks put flesh on the skeletal conventions and how they raised questions about how cabinets should work.
The Australian experience
Initially prime ministers kept their own notes of meetings and left ministers to tell their departments what cabinet had decided. In the 1914–18 war this process did not always work well, but it left Prime Minister Billy Hughes free to determine what he wanted to do too. Only with the highly organised Prime Minister Stanley Bruce did the system change. In 1927 there was the first formal step to create a cabinet secretariat and a set of rules. On 28 November cabinet appointed Sir Neville Howse, a minister, to carry out the duties of secretary to cabinet. Ministers were circulated with a letter describing how cabinet should work and what the expectations on them were.
Officials in the prime minister’s department had prepared a draft. The prime minister then went through the document amending it where necessary and initialling each paragraph when he was satisfied. This first set of rules, like every other one as the procedures developed, were part of the prime minister’s prerogatives to determine when and how the cabinet should be run.
The variations between the draft from officials and Bruce’s final wording illustrate how the official desire for order clashed with the political need for cohesion and flexibility. Bruce added a clause to allow ministers to tell the secretary if they wanted to ‘absent themselves from a cabinet meeting’. Officials wanted due notice for all submissions. Bruce decided to waive these demands ‘in matters of extreme urgency’ but required that ministers advise the secretary if they wanted to bring up an item without notice (see Weller 2007).
Officials wanted tight Treasury control, proposing that: ‘The Secretary will not place questions on the list of waiting subjects until they have been discussed in their financial aspects between the Department concerned and the Treasury’. Bruce softened the stance: ‘It would greatly facilitate Cabinet business if, before submitting questions for Cabinet decision, they have been discussed in their financial aspects between the department concerned and the Treasury.’
The prime minister approved the agenda for each meeting. The secretary summoned ministers and kept records that ‘should be as short as possible and apart from the decision itself will be limited to such explanation as is indispensable to render the decision intelligible’. The secretary would communicate the decisions to ministers. ‘The duty of the Secretary will end after he has communicated the decision to the Minister.’ The secretariat had no responsibility to follow up the decisions.
The draft from officials emphasised that ‘[n]o mention will be made of the views of individual Ministers and the principle of the collective responsibility must be carefully maintained’, but also suggested that, ‘[i]n exceptional cases, by decision of Cabinet, a statement by a Minister may be reproduced and annexed to the minutes of decisions after the approval by the Prime Minister’. Bruce excised the second proposal; there were to be no alternative or individual views after cabinet made a decision.
The official draft and the version approved by the prime minister illustrate the inevitable tension that existed in determining the workings of cabinet. Officials want, indeed need, a good record of what is considered and decided as a necessary basis for implementing the decisions of government. It is useful to know what cabinet decided! Ministers need a forum where they can have a healthy debate, secure in the knowledge that the differences of opinion expressed there will not leak into the media. So any formal records that identified division is unwelcome. Bruce’s amendments reflected this tension. It also illustrated that, whatever drafts the officials may develop, the prime minister decided what will be acceptable, and in what way it will be enforced.
The first version of the rules was short. Gradually over the years additions were made as problems developed and officials responded by amending the rules to prevent a repetition. In 1941 for the first time the secretary of the DPM&C attended cabinet to maintain a record of decisions. The last minister-secretary had proved hopeless, so that departments could not discover what had been decided in cabinet. In wartime this was a problem (for history, see Weller 2007).
In the 1950s Menzies welcomed each new cabinet with a lecture on the need to listen to their departmental secretaries and follow the cabinet rules. Bit by bit, detail was added, on the items that should be brought to cabinet, on format and presentation of submissions. The Cabinet Handbook, as it became titled, grew in coverage and detail. It included definitions of collective responsibility and cabinet confidentiality, required the forecasting of cabinet business, established a ten-day rule for the provision of submissions, set out the format that should be followed, allocated responsibility for implementation and established the rules for access to the cabinet documents of previous governments. In all, by 1980 there were 152 paragraphs.
The Handbook remained confidential; there was no need for the public, or the parliament, to know how the cabinet worked. It was implied that cabinet ministers knew what was required; no one else needed to know. In 1982, in breach of copyright, the Cabinet Handbook was published in Politics, the journal of the Australasian Political Studies Association. No one cared. When the Handbook was being revised in 1984, officials in PM&C decided that, as an earlier version had been published, they may as well publish the revised edition officially with the Australian Government Printing Service. Since then the cabinet handbooks have been public documents, now available on the PM&C website.
The UK experience
In 1945 Clement Attlee became the first prime minister to circulate rules of procedure to ministers. The document, Questions of Procedure for Ministers (QPM), was a consolidation of prime ministerial directives for general guidance’ (Baker Reference Baker2000: 1–3). It was amended by each of Attlee’s successors over the next decades. It remained secret. In the late 1980s Peter Hennessy wrote to Number 10 Downing Street asking that the United Kingdom follow the Australian precedent and release the most recent version of QPM; he sent a copy of the Australian Cabinet Handbook with the request as evidence of precedent elsewhere. He argued that QPM was the ‘nearest thing we have to a written constitution for the proper conduct of cabinet government’ (Hennessy Reference Hennessy1989). He received a blunt refusal (for the history of QPM, see Baker Reference Baker2000).
He tried again when John Major became prime minister. This time he had more success. Major agreed in the 1992 election to release the document. He raised the issue in cabinet. It was the only time QPM was discussed by ministers, and the question was whether it should be released, not what procedures ministers should follow. There were some cosmetic changes to the document, making it more positive in places. The core elements were maintained.
The released document was stapled and xeroxed; its somewhat amateur presentation was evidence of its standing as a temporary and working document. QPM became the Ministerial Code in 1997. A number of versions have followed. There were also codes for special advisers and civil servants, all now public and available on the Cabinet Office website.
In recent years the government decided to go further. Prime Minister Gordon Brown asked the Cabinet Office to draft a cabinet manual ‘to be a source of information on the UK’s law, conventions and rules, including those of a constitutional nature, that affect the operations and procedures of government’ (United Kingdom Cabinet Office 2010). Seen as particularly important was the chapter that proposed procedures for settling a situation where no government won a clear majority at an election. The titles of the chapters tell the story. They included: Elections and government formation; The Executive – the prime minister, ministers and the structure of government; Collective Cabinet decision-making; The Executive and parliament; The Executive and the law; and Ministers and the civil service. The manual also covered relations with devolved administrations, the European Union and other international organisations.
Prime Minister Cameron wrote in the foreword to a version that was revised and released after the creation of the coalition government that:
The Cabinet Manual sets out the internal rules and procedures under which the Government operates. For the first time the conventions determining how the Government operates are transparently set out in one place. Codifying and publishing these sheds welcome light on how the Government interacts with other parts of our democratic system
He stated that ‘the laws, conventions and rules do not change from one administration to the next’.
Of course there were still special circumstances. All ministers ‘were bound by the collective decisions of cabinet, save where it is explicitly set aside’ (4.2). Is that a passing rule or a binding constitutional convention? Will it survive a return to majority government?
There were bigger concerns too. Was the manual, or could it become, the first step towards a written constitution? On that there was debate. Committees in both the House of Commons and the House of Lords took evidence and reported on a draft of the manual. The Cabinet Secretary insisted the manual was ‘by the executive, about the executive and for the executive’ (HC 2010–2011). It was a set of internal rules and guidelines, nothing more.
The problem for members of parliament was that in seeking to codify and record conventions, the manual also had to interpret them. The draft was littered with phrases like ‘usually’ and ‘normally’. But a convention was ‘defined as “rules of constitutional practice that are binding in operation but not in law”.’ There was a clash in the idea that a convention was said to be binding and yet could not be defined with precision.
There is certainly recognition that the manual would create expectations of proper behaviour. The cabinet secretary stated at one point in his evidence: ‘The question is what should be the expectations about what PMs should do? Prime Ministers will always be able to do whatever they want in these circumstances. The convention says what they should do’ (HC 2010–2011: 23).
Here lies the challenge: determining what they should do while acknowledging that they are not bound to do so. The manual sets out normative understandings of what was proper, even if political calculations might sidetrack them.
The House of Commons discussed whether the manual should be debated in parliament and then endorsed. It agreed with the sentiment of the House of Lords Committee that such a step would give it a false status in its current form as a document by and for the executive. But the committee believed that if the manual were to become part of a shared understanding of the way that the UK’s previously un-codified constitution should develop, the parliamentary debate would be appropriate.
The significance of conventions, guidelines and rulebooks
There are a number of conclusions that can be drawn from the above discussion.
First, all the cabinet rules, handbooks, guidelines and codes of conduct are essentially the prime ministers’ prerogative, to change, alter and interpret. That is true of all the countries mentioned. The prime ministers sign off on the documents. After each election, and particularly when there is a new government, the prime ministers have the opportunity to revise and re-issue the instructions. Even when the prime ministers endorse the rules in a foreword, flexibility remains.
Of course the drafting is done within the bureaucracy, and officials will at times add sentences or sections that cover problem areas that have emerged in the past few years, whether they relate to the procedures and content of submissions or the behaviour of ministers. The ever-expanding content, and the new documents, make up a secret history of past errors and crises. Officials want to make procedures predictable and comprehensive. On the other hand, the prime ministers want to be persuaded that the new rules of procedure are effective and will work to encourage ministers. The need to maintain a balance between political and administrative imperatives will continue. Decisions on the content and format of the rules will be made by prime ministers. The rules are theirs alone. The only occasion when QPM were open for discussion within cabinet was when Major wanted to discuss their publication, not their content.
Prime ministers too decide when and if the rules will be enforced. If there is a ten-day rule to require that ministers have the opportunity to consider issues, prime ministers can decide to dispense with that rule. If items should go first to a committee, prime ministers can still bring them directly to cabinet. If the practice is to circulate decisions to all ministers, prime ministers can decide to limit that circulation. Cabinet committees can be given the authority to make final decisions. Sometimes prime ministers will announce a policy before cabinet has had time to consider it. If they then bring it to cabinet for endorsement, ministers are rather bound to support the prime ministers, as the alternative is to effectively censure and remove them. This hasn’t ever happened.
None of these powers are new. Prime ministers used these tactics long before they were incorporated into handbooks and codes. They remain a matter of calculation. What is different is that there are sets of basic rules from which judgements can be made about the wisdom of the judge.
What would happen if they became more than that? That is a very British question. When a committee of the House of Commons reviewed the cabinet manual, the question was raised whether it was the first step in codifying the UK Constitution; by writing the conventions down, the manual created expectations of proper behaviour. There was a concern that ‘a cluster of guidelines can grow and harden, first into expectations, then into conventions and ultimately into code if not quite into a fully-fledged constitutional artefact’ (Hennessy Reference Hennessy2000). If its application was to become broader, should the House of Commons become more formal in its endorsement? The cabinet office insisted not; it was, they said, ‘of the executive and for the executive’ (HC 2010–2011).
Indeed the questions raised issues about the distinctions between entrenched constitutions, legislative limits on the executive, regulations and codes/guidelines. The House of Commons were concerned that the manual might move up that scale to become formalised, even legislated. Yet that was never an issue in Australia. Given an entrenched constitution it was always clear there that these were indeed rules for the running of government, creating expectations perhaps, but never more. The manual developed and explained rules regarding how ministers should behave but did not provide a legislative basis for their actions. That already existed in the Constitution, but only in the barest of forms. There was in practice little difference between a cabinet within a constitution and one where ideas are unwritten; but there was a greater angst about possible directions. No Australian would suggest, as Peter Hennessy did, that QPM represented elements of the UK Constitution. In countries with a written constitution, there has never been such a concern. Cabinet handbooks are guidelines, useful, creating expectations indeed, but never seen as anything more.
As soon as it becomes more, then the question is raised as to whether it is a document that is open to be justiciable. Can a case be made in a court that the ‘proper conventions and procedures’ have not been followed? There remains a common conviction that cabinet debates should be private. Insofar as it dealt with cabinet, this made good sense. Cabinet discussion is a debate among colleagues who at least had the future of the government in common. Outside the room they want to appear united; inside they could disagree until they came to a decision. Cabinet developed to meet a need: the opportunity to have a frank and open discussion about difficult problems, where alternative solutions and policy options were possible.
If those protections were removed, if cabinet became more open, ministers would simply remove their discussions somewhere else and would use the formal cabinet meetings to endorse without debate what they had decided already. Cabinet would very quickly become a formal and dignified, rather than a working, part of the constitutions.
That was behind the Australian cabinet decision to protect the cabinet notebooks from courts. The notebooks were the records kept by the official note takers. They were not a verbatim record of the discussions, rather they included as much of the discussions as the individual note taker thought necessary to ensure an accurate decision. Some were fuller than others; all were in part idiosyncratic. When a court sought to subpoena the records cabinet passed legislation that protected them for fifty years. They may in any event have not proved useful, but it was important that individual minister’s views were not publicised.
Cabinet is valuable because of its flexibility. Prime ministers can adjust the working and the forums in the way that they see best. They do not have to ask if the procedures are somehow right, but rather whether they work in reaching a properly informed decision.
Are cabinet rules conventions?
The introduction to this volume cited Wheare’s (Reference Wheare1960) definition of a convention as ‘a binding rule, a rule of behaviour accepted as obligatory by those concerned in the working of the Constitution.’ Are the rules discussed here binding or accepted? The cabinet manual talks as though UK conventions are binding and clear, even while qualifying them. Therefore the answer must surely be a ‘yes, but…’
Collective responsibility may be accepted as a binding rule, but quite what it means and how/whether it will be applied varies from case to case. Ministers take the concept of ministerial responsibility seriously. They know that they have to account for their actions and answer to parliament. But they do not know exactly what is regarded as satisfactory performance and whether, in a time of crisis, they may fall short of that mythical line of competence. Nor can they predict whether prime ministers will stick with them or cut them adrift if they perform poorly or get caught in a scandal.
So too the ‘rules’ of cabinet are less than binding. Even the most solid of ‘conventions’ may suddenly be found short. All the handbooks have a clause that limits the access to the cabinet records of previous governments. Access to those records requires the approval of the leader of the party that had held government. That was the ‘rule’. But in 2014 the Australian government released some cabinet records of its predecessor to a royal commission without asking approval. There was nothing the former government could do to prevent it.
That decision might have been an extreme. It raises the question whether there are any sections of the cabinet handbooks or codes that could not, with the approval of the prime minister, be put into abeyance. If the prime ministers are the author, judge and executioner for the application of the handbooks, then at any time they can chose to ignore a rule. At different times they have bypassed, ignored or interpreted some of them almost out of existence. The cited comment of the UK Cabinet Secretary that prime ministers will do what they want merely stresses that interpretation.
But they still always do so while insisting that their interpretation aligns with the principles of collective or ministerial responsibility. The principle is usually largely kept intact as an element of political necessity. To ignore any one rule is surely very different from saying prime ministers can put all the rules into abeyance. Nor would they want to; they are essential elements of cohesive government. If it is difficult to argue that the rules and conventions of cabinet are prescriptive and precise in determining how prime ministers can act, but that is not to suggest they can ignore all the rules at the same time.
Of course some argue that is exactly what has happened on occasion. It comes to a question of what cabinet government constitutes. One view can be termed constitutional. Blair, it was argued, practised ‘sofa government’, because cabinet no longer made the crucial decisions; rather these were determined in committee and through cooperation between the prime minister and committee chairs. Cabinet was for an exchange of information; it was no longer the driving force of government. Canadian conclusions were similar: cabinet government was dead under Chrétien (Savoie Reference Savoie1999).
An alternative political or realist view is that modern pressures have simply changed the demands on prime ministers and on cabinet. Conventions have adjusted to satisfy these new demands. One academic reports an early comment from Blair that ‘Ministers have to understand they are the agents of the centre. They have been sent to the departments to carry out a strategy.’ His evaluation of Blair’s statement is that: ‘I cannot imagine many other Prime Ministers saying that’ (Kavanagh, evidence to 2010 HL 30: 22).
This realist version wants to describe what is and then make it work better. Blair adviser Jonathan Powell suggests that: ‘Rather than arguing about the death of Cabinet government, when it in fact died a long time ago, we should spend more effort reinforcing the cabinet committees and their supporting infrastructure as a key part of government decision making’ (cited in HL 2010: 180).
A civil servant, now cabinet secretary, too shrugged off the debate:
I think it partly depends on whether you mean Cabinet government as a meeting once a week or Cabinet government as a set of principles… Every single decision coming through one meeting a week at which there are 20–25 people is not a test of whether Cabinet government is dead
He suggested that civil servants tried to ensure that ‘all the Cabinet departments and Cabinet Ministers with a responsibility have every opportunity to debate, discuss, disagree, agree and we do not announce a policy unless everyone with an interest is signed off and then everyone is bound by the principle of collective responsibility [that is] alive and well in Whitehall’ (HL 2010: 158).
If that interpretation is accepted, then it may be possible to argue that cabinet government can be seen as an only partly defined convention, and that under pressure none of the component parts can be regarded as binding. If government is collective government, and the consent of the collectivity of ministers, or at least their acquiescence, has to be obtained, then there are a multiplicity of ways that can be achieved. Handbooks and manuals are one interpretation, albeit the current one. How seriously they are regarded (and the evidence suggest they are taken more seriously in the United Kingdom than in Australia), then the costs of ignoring them may be greater. But they still can be adjusted for new circumstances.
Prime ministers who do not work with their ministers can create trouble for themselves, not so much on individual policy issues, but in the longer run. New prime ministers, whether following Thatcher, Blair, Chrétien or Rudd, promise that they will re-institute ‘proper’ cabinet government; they appeal to a broader notion of how government in these systems should be run.
If that is so, then the political conventions of ministerial and collective government fall short of Wheare’s binding and obligatory rules, because they are too vague in their application to be binding. The handbooks and codes that instruct ministers and explicate these conventions are less than binding too because prime ministers can interpret them on any occasion in a way that suits their purposes. That does not mean that they do not matter or that ministers somehow feel relieved of the imprecise burden that they impose. Flexibility does not define these ideas out of existence. Rather, they need to be understood within the broader convention that prime ministers work with ministers in changing forums designed to achieve desired outcomes. These forums may include cabinet meetings but are not limited to that. Cabinet government is an ideal, a convention, a process, to achieve the government’s outcomes; it cannot be limited to a forum or a weekly meeting.
Perhaps Wheare’s formulation is too rigid. Ministers understand the conventions and their flexibility. Most prime ministers are aware that there are limits beyond which it is unwise to push. The expectations created by the conventions of cabinet government are more important than the precise meanings of the grand conventions or the implementation of binding rules of procedure that determine how cabinet government will work.
