62.The FTPA includes two mechanisms for dissolving Parliament before the end of a five-year term. We consider the operation of each of these in this chapter and the implications of bypassing the FTPA with a ‘notwithstanding’ bill.
63.In doing so, we consider what mechanisms, if any, are appropriate for bringing about an early general election; what threshold should be set for any such mechanisms; whether legislation should make provision for confidence votes; and whether the House of Commons should be required to approve the date for an early general election.
64.Section 2(2) provides that, if two-thirds of all MPs vote in favour of the motion “That there shall be an early parliamentary general election”, the Prime Minister sets the date of the general election, with Parliament dissolving 25 working days before this date. In our report on the Fixed-term Parliaments Bill we concluded that the requirement for a two-thirds majority was appropriate and meant such an early election would most likely require cross-party support from MPs.
65.Subsequent events allow us to evaluate this conclusion. Theresa May had little difficulty in securing the two-thirds majority of MPs required to dissolve Parliament for a general election in June 2017. In contrast, Boris Johnson was unable to obtain the necessary two-thirds majority on three occasions, and was only able to secure the early general election that he sought through a bill to sidestep the FTPA.
66.There was debate amongst witnesses about whether the support of two-thirds of all MPs was an appropriate threshold. Professor David Howarth thought that “Calling an early election should be a matter of broad consensus, not party advantage. The two-thirds requirement achieves that consensus.”
67.Professor Petra Schleiter agreed, noting that the two-thirds majority is “unusual, but not inappropriate.” Some other countries have a lower threshold to initiate an early general election, she said, but this is combined “with further checks, for instance, by placing the ultimate decision to dissolve parliament in the hands of the Head of State (President). Since no such additional checks are used in the UK, a high majority threshold is the appropriate mechanism for checking the government.”
68.Conversely, Vernon Bogdanor, Research Professor at the Centre for British Politics and Government, King’s College London, said that the two-thirds threshold for an early election was “too high. It could mean that a House of Commons in which the majority seek a dissolution is unable to secure it.” Lord Lisvane mooted the possibility of lowering the threshold: “I have seen suggestions that it should be 50% plus one.”
69.Alternatives to either a two-thirds or simple majority were identified by Professor Robert Hazell:
70.There was debate among witnesses as to whether the two-thirds threshold was a meaningful constraint on a government seeking an early election. Some argued that, in practice, it was very difficult politically for an opposition party to refuse an early election to a government seeking one. Events in April 2017 appeared to verify this theory. Labour, as the largest opposition party, whipped their MPs to back the Government’s motion, resulting in a comfortable victory for Theresa May and thus enabling the early election that she desired.
71.The argument that this threshold is a negligible constraint was challenged by the Government’s three failures to achieve the necessary two-thirds support for an early election in September and October 2019.
72.Professor Gavin Phillipson, speaking after the first two unsuccessful attempts to use the provision in 2019, said:
“It is interesting to compare the reaction to 2017, when Theresa May called her election. Then, opponents of the Act … said, ‘That shows that the FTPA is a dead letter. It has changed nothing, because the Opposition will never not be able to vote for an election and therefore it is a whole waste of time’. We have now had these two refusals proving the exact opposite: that the Commons will, at times, choose to refuse a general election. To me, that shows the worth of the Act.”
73.It is difficult to draw conclusions from the unusual circumstances of autumn 2019. In this period the House of Commons rejected holding a general election on three separate occasions before consenting to one through a different route. While the two-thirds threshold can, in certain circumstances, be a meaningful constraint, it can also perpetuate political instability.
75.The second provision in the Act that may result in an early general election is set out in section 2(3)–(5). This initiates the dissolution of Parliament if a Government loses a statutory vote of no confidence in the House of Commons and, in the subsequent 14 calendar days, the Commons does not pass a motion of confidence in the Government.
76.This provision is one of the most contentious features of the Act, leading to debate on the impact it has had on the notion of ‘confidence’ in the Government. Before the Act it was a constitutional principle that a government’s authority was conditional on having the confidence of the House of Commons. Dr Prescott set out the different scenarios in which the Commons might demonstrate it had lost confidence in a government:
(a)“An explicit motion of no confidence;
(b)An amendment to the Loyal Address expressing no confidence in the government;
(c)A defeat on a vote that the government states that it intends to treat as a matter of confidence;
(d)Potentially other votes, such as the budget or the Loyal Address itself could be a matter of confidence … Furthermore, a government could choose to resign following a serious defeat or series of defeats.”
77.If a government lost the confidence of the House of Commons, the Prime Minister had two choices: to resign and allow the formation of a new government or to seek a dissolution from the Monarch and a general election. These options were constitutional conventions and not codified.
78.Since the passage of the FTPA there has been only one use of the statutory no-confidence motion. On 15 January 2019 the Government was defeated in a “meaningful vote” on its agreement for the UK to leave the European Union by 432 votes to 202. The following day the Government made time for a motion of no confidence tabled—using the wording specified in the Fixed-term Parliaments Act—by the Leader of the Opposition. The motion of no confidence was unsuccessful, with the Government winning by 325 votes to 306.
79.Before the Act, the Government would have been able to declare a vote on such a significant policy as a confidence issue—meaning that if it was defeated the Government would resign to allow an alternative to be formed or, more likely, would seek a dissolution from the Monarch. Declaring a matter an issue of confidence traditionally encouraged recalcitrant backbenchers of the governing party to support the Government. Without the prerogative power of dissolution, this was not as straightforward.
80.In subsequent months, the Government lost two further votes on its withdrawal agreement, yet at no point did the House of Commons indicate that it had lost confidence in the Government. This created the unusual situation of the Government possessing the confidence of MPs in a statutory sense, but not in a manner which allowed it to deliver its most significant policy through the Commons.
81.The effect of the Act was distinctly to separate losing a vote on key policy matter from losing a vote of confidence. A Government might still attempt to declare a vote to be a matter of confidence for political purposes, but it can no longer strongarm MPs with the promise of an immediate election. This is because the power to request a dissolution is no longer in the hands of the Prime Minister but requires a separate vote of the House of Commons. It allows MPs to oppose the Government and resist an early general election.
82.We heard wide-ranging views on the extent to which the doctrine of confidence had been altered or damaged by the Act. There was disagreement about how the conventions around confidence, as previously understood, applied following the creation of the statutory process in the FTPA. Professor David Howarth, suggested that the Act had not materially changed the confidence principle:
“The Act does not affect the ability of a government to threaten to resign or to remove the whip from recalcitrant backbenchers or to ask for a dissolution if it loses a specific vote, and so it is difficult to see how the Act could have replaced the doctrine of confidence or even affected the content of the doctrine. For example, the Johnson government treated the vote in the Commons on 3 September 2019 on the Standing Order 24 motion as a matter of confidence and deprived those who rebelled of the whip. It then proceeded to ask for a dissolution by putting down an early election motion.”
83.Philippe Lagassé acknowledged that the Act:
“does not prevent the Government from declaring that a vote or signature piece of legislature is a matter of confidence. Nor does the Act negate the notion that the reply to the Queen’s Speech and money bills should be considered matters of confidence. Likewise, the Act does not prevent the Commons from effectively withdrawing its confidence in the Government in other ways, such as finding it in contempt or preventing the executive from moving ahead with its legislative agenda.”
He went on to say that the Act “indirectly” changed the notion of confidence, as a government could argue that the statutory vote of no confidence had “become the only binding vote of confidence.”
84.Robert Craig argued that the Act had “significantly affected the operation” of the confidence doctrine:
“At one level, the government must retain the confidence of the Commons to continue in government. But at a second and alternative level, the government is, or was before the FTPA, entitled to seek the confidence of the electorate as a whole, directly, and the FTPA prevents that unless two thirds of the Commons consents. The very calling of an election is an important aspect of the doctrine of confidence that has been detrimentally affected by the advent of the FTPA.”
85.Professor David Howarth concluded:
“The link [the Act] creates between allowing an early election and the doctrine of confidence is confusing and should be eliminated. The lack of detail about what happens in the 14-day period provides another reason for simply eliminating the vote of no confidence route to an early election.”
86.Professor Vernon Bogdanor wrote: “The provision in relation to a no confidence vote should be repealed so as to restore the status quo ante.”
87.A government’s authority derives from possessing the confidence of the House of Commons. The Fixed-term Parliaments Act did not change this constitutional principle, but it has clouded the situation. It is now possible for the Government to retain the confidence of the House of Commons in a statutory sense—winning a vote on a motion of no confidence—while having lost it in the political sense of lacking support for a key part of its policy agenda.
89.While the consequences of a vote of no confidence were changed by the Act, the process by which a motion is debated and voted on in the House of Commons remains largely unaffected. Lord Lisvane explained “a motion of no confidence [is] put down by the Official Opposition and typically it will appear as an EDM … Erskine May says that [such] an Early Day Motion is invariably found time [for debate], but in my experience that [happens within] two to three days maximum.” Lord Norton of Louth agreed that “The convention is if [a motion of no confidence is tabled by] the Leader of the Opposition by virtue of being the Leader of the Opposition—in other words the alternative Government—the Government will find the time fairly quickly for confidence to be tested.”
90.In December 2018, following the postponement by the Government of the first “meaningful vote”, members from the smaller opposition parties—but not from the frontbench of the Official Opposition—tabled a motion of no confidence in the Government. No time was allocated to debate that motion. Lord Lisvane set out the rationale:
“On Motions tabled other than by or headed by the leader of the Opposition, there is little assumption that those will be found time on the Floor of the House … There is a logic in that because that is a demonstrative activity. It is very unlikely, were it to be on the Floor of the House, to lead to a possibility of the Government being defeated. Only the Official Opposition is in a position to do that or to orchestrate that.”
92.A peculiarity of the Act is that, in some circumstances, a government might consider it advantageous to table a vote of no confidence in itself. This might arise if a government wanted an early general election but could not obtain the necessary two-thirds majority, was reluctant to seek to pass a ‘notwithstanding’ bill, and saw little risk of an alternative administration being formed.
93.Philippe Lagassé expanded on this hypothetical situation:
“Orchestrating a motion of no confidence to force an early election would face two hurdles: it would either be embarrassing for the Government to present such a motion or the opposition would be aware of what the Government was seeking to achieve and refuse to cooperate. Yet such manipulations of sections 2(4)–2(5) are not beyond the realm of the possible or the plausible.”
94.Dr Craig Prescott thought that: “Such shenanigans may be controversial in the short-term, but this is likely to dissipate during the ensuing election campaign.”
95.In some situations a government could manipulate the no-confidence process to enable an early general election. This would be against the spirit of the Fixed-term Parliaments Act, but perversely the Act makes such a scenario possible.
96.A successful vote of no confidence is unusual. According to the House of Commons Library, “Since 1895, governments have been defeated on questions of confidence on four occasions. The defeats on questions of confidence in 1895 and January 1924 led to the resignation of the Government and the defeats in October 1924 and 1979 were followed by requests for a dissolution.” In each of these instances there was an immediate “event” following defeat. Section 2 of the Act introduced a period of 14 calendar days before a potential dissolution, allowing several different possibilities to arise. Dr Prescott envisaged four such outcomes:
(a)“The government could regain the confidence of the House following negotiations with other political parties.
(b)It is clear that someone else is best placed to command the confidence of the House. Typically, this might be the Leader of the Opposition, whose party has reached an agreement with other parties to support it in office. For example, if the junior partner of a coalition government decides to support the main opposition party.
(c)No agreement is possible between the parties, and the 14-day period elapses without the Commons passing the motion, “That this House has confidence in Her Majesty’s Government”.
(d)66% of MPs vote for an immediate early general election, and so dispensing with the remainder of the 14-day period.”
The range of possibilities illustrates the potential for significant political uncertainty.
97.Section 2(3)(b) provides for a 14-day—calendar days rather than House of Commons sitting days—period following the Government losing a vote of no confidence before the dissolution of Parliament and an early general election. The introduction of a 14-day period as an extra step following a successful vote of no confidence was a change from past practice and convention.
98.Most witnesses considered that 14 days was at the upper end of what might be considered a tolerable amount of time following a vote of no confidence. Mark Ryan thought that the 14-day period could be shortened to “reduce any political chaos or instability which could arise following a vote of no confidence in an incumbent Government. After all, if there is a credible Government in waiting, is the period of 14 days really necessary? Perhaps this period could be halved in order to minimise any political uncertainty.”
99.Former Cabinet Secretary Rt Hon Lord Butler of Brockwell was more sympathetic to having 14 days:
“I would not argue for longer than 14 days because you want the period of uncertainty to be as short as possible. It should be long enough for the parties to try to see if an alternative Government could be formed without a general election, but 14 days ought to be long enough for that. I do see some advantage in there being that period because if you could solve the issue by another Government being formed without putting the nation to the trouble of a general election there is an advantage in that.”
100.We do not envisage many circumstances where a successful vote of no confidence would result in a change of party in government. It is more likely that there would be a dissolution or a reshaping of the existing Government such that it can regain the confidence of MPs.
101.Under the provisions of the Fixed-term Parliaments Act, if it appeared unlikely that a new administration could be formed within 14 days, the two-thirds early parliamentary general election motion provides a mechanism to truncate the period of uncertainty.
102.As Lord Butler explained, one of the issues with the 14-day period is that, “Because the Act has not been tested in that way so far, conventions have not been established.” There are few historical or analogous situations for the key actors in the 14 days—the Government, the opposition parties seeking to form a new government, the civil service and the Monarch—to draw on. For instance, one of the uncertainties during this period would be how the Government might behave after having lost the confidence of the House of Commons. This was a moot point before the Act: a government either resigned or there was a dissolution and the purdah convention commenced. Now, under the Act, there would be a period of political uncertainty, during which the country would continue to be governed by the incumbent ministers, with potentially difficult and contentious decisions needing to be made.
103.Professor Robert Hazell observed that “The caretaker convention applies when a government cannot command the confidence of Parliament: for example, after it loses a formal no confidence motion under s2(4) of the FTPA.” The Cabinet Manual does not explicitly refer to the caretaker convention, but under the heading of “Restrictions on government activity” it states:
“While the government retains its responsibility to govern and ministers remain in charge of their departments, governments are expected by convention to observe discretion in initiating any new action of a continuing or long-term character in the period immediately preceding an election, immediately afterwards if the result is unclear, and following the loss of a vote of confidence. In all three circumstances essential business must be allowed to continue.”
104.Dr Ruth Fox said that the caretaker convention included “decisions of a long-term nature, senior appointments to the Civil Service, [and] spending significant sums of public money.” Dr Catherine Haddon raised concerns about the caretaker convention:
“The principles around caretaker government in this country are still very reliant not even just on constitutional convention but on an expectation of it being customary, which is the phrasing that is used in the guidance, and that is possibly too weak a restriction in those periods.”
105.Professor Petra Schleiter also thought the caretaker convention had shortcomings. She said the caretaker convention principles:
(a)“do not clarify effectively the normative basis for limitations on government power during caretaker periods;
(b)do not stipulate unequivocally that a caretaker government cannot resign until its successor is formed and do therefore not fulfil the minimal function of all caretaker conventions, i.e., to secure the continuing existence of a government at all times;
(c)do not clearly specify when caretaker situations terminate; and
(d)lack clarity regarding the nature of restrictions on the executive during caretaker periods and their enforcement.”
106.It is unknown what might happen in the 14 days following a government losing confidence in the House of Commons. But, for the avoidance of doubt, the Cabinet Manual should state that the existing Government remains in office until the Monarch invites a new Government to be formed. This remains the case even if the Government has lost a vote of confidence in the House of Commons. The Prime Minister would not be “squatting” in Number 10 in such circumstances; rather, they would be doing their duty in maintaining Her Majesty’s Government until such time a replacement was appointed.
107.The only provision in the Act about the formation of a new Government during the 14 days is section 2(5). This stipulates that there will not be an early general election if a majority of MPs support the motion “That this House has confidence in Her Majesty’s Government.” It is possible that this could be the incumbent Government regaining the confidence of the House of Commons, which would provide it with the authority to return to governing free from the constraints of the caretaker convention. Alternatively, it is possible that the party or parties of Government might change.
108.One issue with the latter possibility is how an alternative administration could demonstrate to the Monarch that it is best placed to become the next Government. Dr Ruth Fox explained that there is a “procedural gap” in the Act as to what happens in the 14 days:
“we do not have procedures in place to be clear about exactly how the identity of an alternative Administration would come about … the problem is if there is a potential Administration and it is not exactly clear who it might be, there needs to be clarity on the part of Buckingham Palace in terms of next steps. It would have been better if, for example, the House of Commons, knowing that these provisions were in the Act, had looked at this and determined what the range of options might be, if needed, in order that the procedural rules of the game are clear before they had to be confronted and dealt with once the circumstances arise.”
109.Lord Butler suggested that the lack of clarity on sequencing had led to a “Catch-22” situation: “You cannot be Prime Minister unless you can get a vote of confidence, but you cannot get a vote of confidence until you have been asked to be Prime Minister. It is difficult to resolve.”
110.Witnesses offered different approaches as to how a potential alternative administration might use House of Commons procedures to provide clarity to the Monarch. Dr Fox thought that there were two possible options: “One option is indicative votes. Another option might be something such as an Early Day Motion signed by the MPs who would support an alternative Administration led by X.”
111.Lord Lisvane was hesitant about using an early day motion (EDM) for this purpose:
“I would be very uneasy about an EDM. We all know about the authority of signatures to an EDM. They can be put down. You can go into the Table Office and ask them to take your name off. Casting a vote, on the other hand, is a different matter … There has to be a means, therefore, by, and again I hesitate to use this phrase, an indicative vote of demonstrating that a certain grouping will command a majority.”
112.The Fixed-term Parliaments Act is silent on the process for finding an alternative administration during the 14 days after a vote of no confidence, save for the provision for a vote of confidence. The Act may create a circular situation whereby an alternative administration would need to demonstrate a majority in support in the House of Commons, but may not be able to demonstrate said support until it is in office.
113.The Cabinet Manual states:
“In modern times the convention has been that the Sovereign should not be drawn into party politics, and if there is doubt it is the responsibility of those involved in the political process, and in particular the parties represented in Parliament, to seek to determine and communicate clearly to the Sovereign who is best placed to be able to command the confidence of the House of Commons. As the Crown’s principal adviser this responsibility falls especially on the incumbent Prime Minister, who at the time of his or her resignation may also be asked by the Sovereign for a recommendation on who can best command the confidence of the House of Commons in his or her place.”
114.It is a cardinal principle that the Monarch should not be drawn into the political debate regarding who should govern. There should therefore be a process for MPs to demonstrate who is best placed to command a majority in the House of Commons in order to signal to the Monarch who should be invited to form a government. It is for the House of Commons to decide what process would be appropriate.
115.Under the current system a Prime Minister opposed to any alternative administration could block its formation by preventing any effort to command a majority and waiting out the 14 days until an election was triggered. Such behaviour would be deprecated, but it underlines the need for greater certainty on this matter.
116.There is a possibility that an incumbent, caretaker Government may refuse to resign, even if it becomes apparent that an alternative administration can be formed and is better placed to command a majority in the House of Commons. Such a Government could argue that it is preferable for the government to change through a general election after the 14-day period. Depending on the circumstances, a Government behaving in this way may be acting unconstitutionally. The ultimate safeguard in such an instance remains the Monarch who, as Philippe Lagassé said, retains the prerogative to dismiss a Prime Minister:
“This reserve power ensures that the Queen can remove a Prime Minister who is incapacitated or involved in criminality or flagrant unconstitutional behaviour. Yet this is not a power that the Queen would normally be expected to exercise. It remains in place to address exceptional circumstances and situations where other actors, such as the Cabinet, the Commons, or the courts, cannot provide a remedy, and the Prime Minister is either unwilling or unable to act honourably or according to the requirements of the constitution.”
117.After losing a vote of no confidence a Prime Minister must continue to follow constitutional practice. If an alternative Government can be formed during the 14 days prescribed by the Fixed-term Parliaments Act, the incumbent has a duty to resign. A Prime Minister who did not do so would be behaving unconstitutionally.
118.During this inquiry the Government introduced, and Parliament passed, the Early Parliamentary General Election Act 2019. This approach was possible because the then minority Government had the support of opposition parties for an early election. A minority administration without such support would risk the bill being defeated or amended in ways it considered undesirable. The Early Parliamentary General Election Act bypassed the “fixed term” and “early parliamentary general election” provisions of the Fixed-term Parliaments Act. It allowed for an early election without having to achieve a two-thirds majority of MPs or use the no-confidence procedure.
119.A short ‘notwithstanding’ bill has always been a possible way of circumventing the Fixed-term Parliaments Act and was envisaged by some witnesses in advance of it being used. Professor David Howarth did not see a problem with such an approach, arguing that broad consensus rather than party advantage was necessary for an early election and that the passage of a notwithstanding bill would demonstrate that consensus.
120.However, Dr Haddon said that this was:
“another flaw in the Act … the possibility of the Government bringing in a short notwithstanding Bill that just sets it aside. Once again, this goes back to the tube of toothpaste. Once it is open—and Governments are increasingly aware that they could do this themselves if they have a majority—they might just start setting it aside all the time.”
121.Professor Robert Blackburn, Professor of Constitutional Law, King’s College London, said that the Early Parliamentary General Election Act 2019:
“has now driven a coach and horses through the Fixed-term Parliaments Act and as a matter of process is to be greatly regretted. Whatever one thinks of the Fixed-term Parliaments Act, while it is in force its terms should be complied with until and unless it might be amended or repealed. The subject of election timing is fundamental to our democracy and bypassing constitutional law in this manner is akin to disrespecting other legislation of a fundamental character such as the Human Rights Act or Scotland Act.”
58 The Act originally provided for dissolution 17 days ahead of a general election. This was later amended by the Electoral Registration and Administration Act 2013 to 25 days.
59 Constitution Committee, (8th Report, Session 2010–12, HL Paper 69), para 102
60 Written evidence from Professor David Howarth, University of Cambridge ()
61 Written evidence from Professor Petra Schleiter, University of Oxford ()
62 Written evidence from Professor Vernon Bogdanor, King’s College London ()
63 (Lord Lisvane)
64 Written evidence from Professor Robert Hazell, University College London ()
65 Written evidence from Carl Gardner, BPP University of Law ()
66 On each occasion Hansard recorded that the “Question [was] accordingly agreed to, without the majority required under the Fixed-term Parliaments Act 2011.” See HC Deb, 4 September 2019,
67 (Professor Gavin Phillipson)
68 Written evidence from Dr Craig Prescott, University of Winchester ()
69 HC Deb, 15 January 2019,
70 HC Deb, 16 January 2019,
71 Written evidence from Professor David Howarth, University of Cambridge ()
72 Written evidence from Philippe Lagassé, Carleton University ()
74 Written evidence from Robert Craig, London School of Economics ()
75 Written evidence from Professor David Howarth, University of Cambridge ()
76 Written evidence from Professor Vernon Bogdanor, King’s College London ()
77 As the language of the no-confidence motion is set out in the FTPA, the motion is subject to House of Commons Standing Order No. 16 and therefore debate is limited to 90mins unless a prior Business of the House motion makes alternative provision. Prior to the FTPA, no-confidence motions would not have been subject to this Standing Order.
78 (Lord Lisvane)
79 (Lord Norton of Louth)
80 (Lord Lisvane)
81 Written evidence from Philippe Lagassé, Carleton University ()
82 Written evidence from Dr Craig Prescott, University of Winchester ()
83 House of Commons Library, Confidence Motions, Briefing Paper, , 14 March 2019
84 Written evidence from Dr Craig Prescott, University of Winchester ()
85 Written evidence from Mark Ryan, Coventry University ()
86 (Lord Butler of Brockwell)
87 (Lord Butler of Brockwell)
88 Written evidence from Professor Robert Hazell, University College London ()
89 Cabinet Office, The Cabinet Manual, October 2011, para 2.27: [accessed 30 July 2020]
90 (Ruth Fox)
91 Written evidence from Professor Petra Schleiter, University of Oxford ()
92 Fixed-term Parliaments Act 2011,
93 (Dr Ruth Fox)
94 (Lord Butler of Brockwell)
95 (Dr Ruth Fox)
96 (Lord Lisvane)
97 Cabinet Office, The Cabinet Manual, October 2011, para 2.9: [accessed 30 July 2020]
98 Written evidence from Philippe Lagassé, Carleton University ()
99 Written evidence from Professor David Howarth, University of Cambridge ()
100 (Dr Catherine Haddon)
101 Written evidence from Professor Robert Blackburn, King’s College London ()