Written evidence from Raphael Hogarth and Dr Catherine Haddon (FTPl7 )
Public Administration and Constitutional Affairs Committee Fixed-term Parliaments Act 2011 inquiry
Summary of key points
- The system for calling early elections which replaces the FTPA should be clear, should contain protections against abusive exercise of the dissolution power, and must resolve the most troublesome problems of the FTPA. The new system should also respect constitutional principles. Those are the criteria against which government proposals should be judged.
- Going back to the pre-FTPA system would be clear insofar as it would be familiar, but there would remain significant uncertainty about the circumstances in which the monarch can or should refuse a dissolution. Though in theory the monarch would operate as a check on the power under this system, refusal by the monarch may be considered illegitimate in a modern democracy. The monarch will therefore always be highly reluctant to refuse, making the check increasingly unreal.
- If the government does wish to go back to the status quo ante, it should consider adding an obligation on the prime minister to hold a non-binding parliamentary vote on dissolution before he seeks a dissolution from the monarch. If he lost it or failed to win it by a convincing margin, that could provide some democratic legitimacy to a refusal by the monarch and so strengthen the check.
- The government could take the monarch out of the picture by making dissolution a simple statutory power, but this would contain no protections against abuse and so would be constitutionally backward.
- Better than a pure ministerial power would be one which could only be exercised if a majority in the House of Commons backed an election. This system would make it less likely that the courts would ever get involved in dissolution and would help to avoid abusive exercise of the dissolution power by the executive. It would also solve many of the problems with the FTPA. The chance of a 'zombie parliament', which would give the prime minister neither his business nor an election, would be reduced but not eliminated.
Introduction: the objectives and options
- The Institute for Government is a politically non-partisan think tank whose mission is to make government more effective. In this submission, we focus our evidence on the question of what system for calling early elections should replace the FTPA.
- Whatever system replaces the FTPA should be measured against the following criteria:
- The system should be clear and well understood by constitutional actors. The legislation must leave no uncertainty about who has the power to call an election, who can veto it, the timetable for an election thus called, and the roles played by parliament and the monarch. Any ambiguity as to the process of calling an election or dispute over whether an election had been called would risk leading to a decision maker who is supposed to be politically impartial, such as the Speaker or the court, having to make a highly political decision on whether there was to be an election or not. This would be damaging to those institutions and to trust in democracy.
- The system should contain protections against abuse. Any power, if unchecked, can be exercised abusively. Dissolution is no exception, as much as we might hope that those involved will act proportionately and rationally. An abusive exercise of the dissolution power might be, for instance, a dissolution immediately after an election when no party has won a clear majority, or during a moment of intense national crisis (such as a declaration of war). It is also arguable, albeit more controversially, that dissolution is less legitimate if an incumbent government dissolves in order to hold an election when another potential prime minister in the Commons is capable of governing with a working majority. The pre-FTPA system and the FTPA system both contained some protection against abuse; the new system should do so as well.
- As far as possible, the system should solve the problems of the FTPA. Both main parties promised in their manifestos to repeal the FTPA, presumably because they think there is something wrong with it. The government should clearly identify the problems it perceives with the FTPA, and explain how the new system will solve them. Otherwise, this will be reform for reform's sake.
- More generally, the system should respect important constitutional principles and conventions. In the Westminster system, these are important as rules of law to the functioning of democracy. They include the principle that the monarch should not be forced to make partisan political decisions, and the principle that the prime minister derives their authority from the confidence of the House of Commons. Relatedly, they include the convention that the Queen appoints as prime minister the person best placed to command the confidence of the House.
- In our view, there are three main policy options for the government and parliament to consider, which we address in turn.
- The first option is the status quo ante: a return to a system in which the prime minister asks for a dissolution from the monarch.
- The second option is a pure ministerial power: a system in which the prime minister has the power to call an early election himself, without the involvement of the monarch or parliament.
- The third option is a revised parliamentary power to call elections: a system in which the early dissolution power remains in the hands of MPs, but with some modifications to the existing FTPA system.
Option 1: The status quo ante
Whether restoring the status quo ante is possible, and how to do it
- The first question about restoring the status quo ante is whether it is legally possible to do so. Prior to the FTPA, the power of dissolution was a prerogative power. It has been suggested that the prerogative power of dissolution cannot be revived, as the FTPA abolished it forever.
- In our view, if parliament enacted legislation which provided that the FTPA was repealed and that the prerogative of dissolution was revived, then in any event a court would be likely to hold that the prime minister had the power to seek a dissolution from the monarch and the monarch had a power to grant one. The alternative-to hold that the enactment was entirely without effect-would be obviously contrary to parliament's intention in enacting the repealing Act. If the legislation which repealed the FTPA explicitly revived the Septennial Act, as amended by the Parliament Act 1911, then the effect would be that the parliament lasted for five years; if the repealing legislation did not do that then the effect would be that the present parliament lasted forever, until parliament legislated to end it. Neither of these outcomes would be in line with parliament's clear intention in purporting to revive the prerogative of dissolution.
- The only true question, in that case, would be whether the power contained in the repealing Act was most accurately analysed as the old prerogative power revived, or a new statutory power created. In today's constitutional landscape, that does not matter very much. It might be thought that, if the prerogative power has been revived, the courts are less likely to interfere with dissolution, on the basis that, as Lord Roskill held in Council of Civil Service Unions v Minister for the Civil Service  UKHL 9, "prerogative powers such as[ ... ] the dissolution of Parliament and the appointment of ministers as well as others are not [... ] susceptible to judicial review because their nature and subject matter are such as not to be amenable to the judicial process."
- However, in our view that is not a good argument for using the word 'prerogative' in the legislation, for two reasons.
- First, Lord Roskill's dictum now appears dated. The Supreme Court held in R
(Miller) v The Prime Minister and Cherry v Advocate General for Scotland 
UKSC 41 that the question "whether a prerogative power exists, and if it does, its extent [... ] undoubtedly lies within the jurisdiction of the courts and is justiciable". Though the Supreme Court acknowledged that the "exercise" of some prerogatives is not challengeable before the courts on traditional grounds of judicial review, in truth the judgment provides an easy route around that exclusion for a court minded to review the exercise of a prerogative power: to hold that the power does not extend to what the minister has done with it, or that the minister did something which was outside the "legal limits" of his power, rather than the minister exercised his power in an unreasonable or unfair way. The Miller/Cherry case, though certainly controversial in many political and legal circles, is now binding authority on future courts. In the aftermath of that case, therefore, it is unlikely to make much of a difference to the reviewability of dissolution whether the legislation calls the dissolution power a prerogative power or not.
- Second, at least in the short term, using the word 'prerogative' could generate its own legal challenge: namely, if someone tries to argue before the court that revival of the prerogative was impossible.
- Therefore, we conclude that revival of the status quo ante is possible. For the sake oflegal certainty, it would be marginally preferable to do this with ordinary statutory powers than by trying to revive the prerogative.
Whether restoring the status quo ante is desirable
- The old system has the advantage of clarity, insofar as it is familiar and we know how it works. However, there was ambiguity over confidence conventions in the old system just as in the new - in particular, over whether a prime minister was obliged to resign following a vote of no confidence. There was also some ambiguity over the circumstances in which the monarch could, and should, refuse a prime minister's request for a dissolution. In our view, any attempt to remove those ambiguities in the repealing legislation is likely to end in tears. Just as it would have been difficult, before last year, to imagine all the circumstances in which a government might want to prorogue, it is difficult now to imagine all the circumstances in which the government might want to dissolve, or might face a vote of no confidence. A binding legal code on refusing dissolution or on what follows a vote of no confidence could render these aspects of the constitution unnecessarily inflexible. There is nevertheless a case for writing into the Cabinet Manual some discussion of the conventions at play, if the government opts for this option.
- Any system should also contain some check on the exercise of the dissolution power, to ensure that it is not exercised abusively. In our view, this is the criterion against which the "status quo ante" option performs worst. The present check on the dissolution power is the need to secure a super-majority in parliament (or a majority for a one-line bill to circumvent that requirement); the status quo ante check is supposed to be the discretion of the monarch to refuse a dissolution if one is sought in inappropriate circumstances. However, this is an increasingly unreal check. In theory, the monarch can still lean on the
conventional rules known as the 'Lascelles Principles', that she is entitled to refuse a dissolution if the existing Parliament is still "vital, viable, and capable of doing its job", an election would be detrimental to the national economy and she can find another prime minister who could "govern for a reasonable period with a working majority in the House of Commons". However, in our view it is becoming far more difficult, in a 2Pt-century democracy, for a hereditary monarch to refuse a dissolution, possibly even an abusive one. To do so would expose the monarch to allegations of political interference of an undemocratic nature, even if the intention of the refusal was to preserve the good functioning of democracy. The prorogation saga showed that the Queen felt unable to dissent from the advice of her ministers, even in circumstances where there was a strong constitutional case for her to do so. That left the Supreme Court seeing itself as the last possible check against, in its words, "the risk that responsible government may be replaced by unaccountable government: the antithesis of the democratic model". So it intervened.
- One possible modification to the old system, which would go some way to solving this problem, is to require the prime minister to hold a non-binding vote on a dissolution motion in the House of Commons before seeking a dissolution from the monarch. (The legislation would need to specify whether it would still be required if the House were not sitting or prorogued.) The idea would be that, if the three criteria in the Lascelles Principles were met, then the Commons would be more likely to vote against dissolution, in which case the perceived and actual democratic legitimacy of the monarch refusing the prime minister a dissolution would be significantly bolstered. We believe this compromise between executive and legislative roles in dissolution would be a useful addition to the previous system, if the government does seek to return the power to the prime minister.
- As to whether the status quo ante option solves the most glaring problems of the FTPA, its greatest virtue is that it is likely to do so. Insofar as the FTPA risks creating a so-called 'zombie parliament', where the Commons refuses to give the prime minister his business but also refuses him an election, the status quo ante annuls that risk by allowing him to dissolve off his own bat. Insofar as the FTPA generates additional uncertainty concerning what happens in the 14-day statutory period that follows a vote of no confidence, the status quo ante gets rid of that uncertainty by getting rid of the period. Insofar as the FTPA has created confusion (and sometimes wilful confusion) among politicians, particularly those in government, about what constitutes a matter of confidence, the status quo ante is likely to get rid of that confusion, as there would no longer be a special category of statutory confidence motions, distinct from other matters of confidence like the budget and the Queen's Speech.
- On the status quo ante, we conclude that it would solve the main problems with the FTPA but, in its traditional form, would involve insufficient checks and balances on the exercise of the dissolution power.
Whether the legislation should contain an ouster clause
- The government is obviously worried about the courts getting involved in constitutionally important matters, in the aftermath of the Miller/Cherry case last year. The government may therefore be tempted to include in the legislation an 'ouster' clause, which attempts to oust the power of the High Court to judicially review the government's decisions on this issue, providing that the prime minister's decision to seek a dissolution from the monarch cannot be questioned in court. In our view, an ouster clause of this kind would be asking for trouble.
- First, it would be unlikely to work. The courts abhor ouster clauses and have consistently refused to read them literally. Giving the lead judgment in R (Privacy International) v IPT  EWCA Civ 1868, a case about a clause which tried to oust the High Court's jurisdiction to judicially review decisions of the Investigatory Powers Tribunal, Lord Carnwath held that "the ultimate safeguard of judicial review remains essential if the rule oflaw is to be maintained". To include an ouster clause would be asking for a fight with the courts which no one should want to have: inviting a judge to find their way around an enactment in primary legislation with massive political ramifications. It would be better to leave it out.
Option 2: Pure ministerial power
- The monarch could be written out of the dissolution process entirely, giving the prime minister a straightforward statutory power to dissolve. This option would be clear, insofar as it would remove any ambiguity surrounding the circumstances in which the monarch was entitled to refuse dissolution, and it would solve the problems with the FTPA for the same reasons that the status quo ante option would. It might also be thought that this option has the advantage of keeping the monarch out of politics.
- However, in our view a pure ministerial power would be less democratic than a system in which the monarch has a role and would be constitutionally backward. There would be no ex ante check against abusive exercise of the power at all. There would be an ex post check, in the form of the courts, but the courts are less well suited to this job than the monarch is, even taking into account the problems of the hereditary monarch exercising this power. Although both institutions are politically impartial, the courts are supposed to think apolitically and the monarch is not. It is well recognised that the monarch is entitled and required to have regard to the political situation when making decisions; the courts are not well equipped to do this and are heavily criticised when they do.
- There would be a further ex post check in the form of the democratic accountability that the prime minister would face in the election that followed dissolution. However, this accountability is inadequate too. First, an ex post check is not as good as an ex ante one. Second, because the constitutional propriety of dissolution is unlikely to be the main issue in an election, subsuming accountability for dissolution into wider electoral accountability dilutes the accountability for dissolution. It might be argued, instead, that the prime minister has a democratic mandate to make his own decision about dissolution as a result
of the previous election he won. The flaw in that argument is obvious. A prime minister derives his democratic legitimacy from the confidence of the House of Commons and not from a direct electoral mandate; there is no guarantee that a prime minister seeking dissolution would have that confidence at all. The leader of a minority government might lose the support of opposition MPs on a matter of confidence, the leader of a coalition might lose the confidence of his coalition partner, and the leader of a a majority government might lose the confidence of his own MPs. A pure ministerial power is, for all these reasons, undesirable.
Option 3: Revised parliamentary power
- There are some advantages to keeping the power of dissolution in parliament, rather than transferring it back to the executive. First, because of parliamentary privilege - in particular, the rule in Art IX Bill of Rights 1689 that "debates or proceedings in Parliament ought not to be impeached or questioned in any court" - if dissolution is a proceeding in parliament (as it is under the FTPA system), dissolution cannot and will not be questioned in court. Second, a parliamentary role protects against an abusive dissolution, for instance, an attempt by a prime minister to dissolve following an election in which he did not get an overall majority. Third, there is an argument from democratic fairness for a system in which parliament calls elections, not least because it is that parliament that is being dissolved. Research has found that, on average, a prime minister gets a vote share bonus of 6% and a seat share bonus of 12% when they pick the election date, increasing the chances that they stay in office by 51% (although, of course, a prime minister with a majority and obedient backbenchers can pick the timing of an election even when the power does sit in parliament).1
- The question is how those advantages of the FTPA system can be retained, while dealing with some of the Act's problems.
- Parliament could enact legislation which reduced the threshold for an early election from a two-thirds majority of all MPs to a simple majority of those voting, as normal, while getting rid of the Act's provisions on confidence motions. In other words, the prime minister would have the power to call an early election subject to passing a motion for an early election in the House of Commons. This would effectively amount to going back to the status quo ante, but having the House of Commons exercise the same function that the monarch used to. This would solve some of the problems with the FTPA.
- First, that reform would go some way to solving the problem of the paralysed or 'zombie' parliament, in which MPs will give the prime minister neither his business nor an election. Moving to a simple majority threshold and abandoning the FTPA confidence provisions would not entirely solve this problem because it would still be possible, in theory, for the Commons to refuse to give the prime minister his
1 Schleiter and Belu, 'Electoral incumbency advantages and the introduction of fixed parliamentary terms in the United Kingdom', The British Journal of Politics and International Relations 2018, Vol. 20(2) 303-322
business and to refuse a dissolution. If the prime minister wanted to designate a vote on a policy matter as a "matter of confidence", then he could threaten to resign or to move a motion for an early general election, but the threat would not be as powerful as the pre-FTPA threat, because any backbencher who opposed the government's policy but did not fancy their chances in an election could, if they thought it was politically palatable, vote against the government in both votes. However, in our view, the position would be somewhat better than it is now. The threshold for getting an election would be lower, so it would be less likely in any event that the prime minister would fail to get parliament's backing for an election if he wanted one. If the Commons refused the government both an election and its support on matters of confidence, then there would have to be an convention that the prime minister was obliged to resign and, if he refused to do so, that the monarch was in extremis be entitled to dismiss him (though in our view, like all aspects of making and breaking governments, this should be a part of the system which is not set out in legislation).
- The second problem we have discussed with the FTPA is the 'problem of disputed confidence' - that there is now disagreement over what constitutes a matter of confidence. PACAC has, in our view rightly, made clear in a previous inquiry that the view that only statutory confidence motions under the FTPA are matters of confidence is distorted and wrong, but it will nevertheless be an attractive view for any government which is losing key votes and would rather not resign. Like the other options, this option involves the abolition of the statutory 'no confidence' route to an election and so would be likely to solve this problem.
- The third problem we have discussed with the FTPA is uncertainty about what happens or ought to happen in the 14-day period that follows as statutory 'no confidence' motion in order to test the confidence of the House and decide who governs. Moving to a simple majority threshold and abandoning the Act's provisions on confidence would solve that problem, because the 14-day period would no longer exist.
- In short, it may be possible to retain some of the benefits of the FTPA while fixing some of its flaws, by making dissolution a ministerial power subject to approval by a simple majority in the House of Commons.
- The government should, in our view, consider the merits of Option 3, a revised parliamentary power of dissolution. Nevertheless, we recognise that the government appears minded to pursue Option 1 and attempt to restore the status quo ante. If it does so, it should consider how to update that system so as to bolster the efficacy and democratic legitimacy of the monarch's function in that system, particularly by considering a non binding parliamentary vote to support the call for dissolution. The government should not, in our view, countenance Option 2, a pure ministerial power. June 2020