Robert Craig, University of Bristol – written evidence (FPA0013)



  1. I give this evidence in my personal capacity as an expert in Constitutional Law having taught the subject for over 20 years variously at LSE, KCL, Bristol University and elsewhere.[1] I have been published in a number of leading journals. I have had two articles cited in the Supreme Court. One was by Lord Reed in the first Miller litigation.[2] The second was by Lord Carnwath in the more recent Privacy International litigation.[3]





  1. In my submission, the Fixed-term Parliaments Act 2011 (‘FtPA’) has had a significant negative effect on the constitution. This paper explores some of those negative ripple effects making some use of the current Brexit situation as an illustrative example. In my view, the FtPA has had actual real world effects on political outcomes in recent months. This evidence should be considered as a précis of an article published in the Modern Law Review in 2018 on the FtPA that was previously requested by the Policy Analyst of the Committee combined with recent shorter contributions entitled What happens after a vote of no confidence in the PM and What could happen next if the government resigns rather than send the letter? for which links can be located in Footnote 4 below.[4] In this note, I address some of the Key Questions raised by the Committee.





Are the mechanisms in the Act to trigger an early general election appropriate? What impact has the Act had on the notion of the House of Commons having "confidence" in a Government? Is it still possible for the Government to make a vote in the House of Commons on a matter of policy a "confidence" issue?


  1. The doctrine of confidence is the beating heart of the political constitution. Adam Tomkins has argued that the doctrine is a ‘simple – and beautiful – rule’ at the ‘core’ of the ‘remarkable creation’ that is the British constitution.[5] A vote of no confidence is the most political moment in a political constitution. It is its apogee. Historically, it would in effect determine whether a government would continue, whether it would be replaced or whether there would be an election. The FtPA has significantly affected the operation of this doctrine both directly and indirectly.


  1. As Bagehot pointed out, the British system is famously ‘fused’. The executive can continue in power only for as long as it can retain the confidence of the House of Commons. The legislature used to continue only for as long as the government did not advise Her Majesty (‘HMQ’) to call an election. Both halves of this equation were balanced and retained ultimate power over the other. The FtPA has upset this delicate balance.


  1. The legislature and executive in the UK are chosen at the same time and in the same election. It is sometimes, erroneously, argued that voters only choose their local MP in a general election. This may be true as a matter of strict law, but this very example neatly illustrates the narrowness of trying to understand a political constitution solely through a legal prism. The ‘local MP only’ approach is sometimes described as the ‘representative government’ conception of the constitution.


  1. In fact, and this is the political reality, voters choose their MP and they vote on the manifestos of the executive slates that put themselves forward as a potential government. Our vote is bifurcated. This is entirely appropriate in a fused system. This contrasting ‘fused’ approach, defended in this paper, is sometimes described as the ‘responsible government’ view of the constitution.


  1. One clear example of the centrality of this conception in the political constitution is the operation of the Salisbury Convention in the House of Lords,. This rule represents a careful compromise between the two Houses in response to the direct and legitimate electoral mandate conferred on the executive by the manifesto on which they stood winning the most votes. Our system, at both macro and micro level, is a ‘winner takes all’ competition and is no less democratic for being thus structured.


  1. In normal times, and with a normal majority, the representative government and responsible government conceptions work in harmony and in harness. The Brexit referendum, and the hung Parliament of 2017, have pitted these two conceptions against each other in extremely unhelpful and damaging ways.


  1. The FtPA is a clumsy and legalistic attempt to regulate the boiling cauldron at the heart of the political constitution. As 2017 showed, and as 2019 appears to be about to show, the FtPA fails even in its most basic, self-described, mission of fixing the term of parliaments. In the end, if the politicians think there should be an election, then an election there shall be.


  1. Some have argued that the FtPA is unfairly castigated because it was not supposed to affect the doctrine of confidence, which underpins governments, but rather was only supposed to fix parliamentary terms, which underpin parliaments. This legalistic attempt to disaggregate government and parliament does not even begin to address the fused nature of the system and unhelpfully disregards the inevitable ripple effects of making the norms surrounding parliamentary terms so much more rigid, with serious knock-on effects on the governments that draw their authority from such parliaments through the doctrine of confidence. It is like trying to pop a balloon half way.


  1. The most serious direct effect of the FtPA, in my submission, is on the self-understandings of the political actors in the system. Senior MPs have given evidence to PACAC in the other place to the effect that, in their view, only a formal no confidence motion under the FtPA triggers the consequences that used to flow from any clear loss of confidence in the government by the House. It has even been suggested that the bringing of a motion of no confidence by the leader of the opposition might not be given time, in accordance with long standing convention, if it did not match the FtPA wording precisely. This is startling. It should be mentioned that the committee report did not endorse this interpretation of the relevant rules and nor does this author.


  1. These developments are unfortunate, to put it mildly, and despite the good intentions of many FtPA supporters, this detrimental impact on long-standing and previously well understood norms perhaps serves as a reminder of what road is frequently paved with such good intentions. It is an unfortunate irony that a measure that was supposedly designed to reduce the power of Prime Ministers could in fact serve to entrench an incumbent in circumstances where, in former times, an election or new premier would otherwise have resulted.


  1. There is one common argument that is regularly wheeled out in favour of the FtPA with which this author strongly disagrees. It is frequently suggested that determining the date of a general election is an unfair political advantage for the Prime Minister (‘PM’). It is sometimes described as handing the starting pistol to one of the competitors. This argument is seriously misconceived, in my submission. To continue the analogy, even if the pistol is in the hands of the PM, there is every chance he or she will shoot themselves in the foot with it, thereby hobbling themselves in the race. Examples include Theresa May in 2017, Ted Heath in 1974 and Gordon Brown’s change of heart in 2007 which was arguably grounded in precisely his fear of such a decision going wrong.


  1. General elections cause inconvenience to millions of people, including Brenda from Bristol. The idea that the power to call an election is somehow a ‘party political’ advantage is extremely questionable. Indeed, such a view arguably suggests a suspicion of politicians, normal politics and the political constitution itself from those who would prefer a more rigid and legalistic approach. It is slightly odd to criticise the power to call an election for being ‘political’ in nature. Of course it is political. Everything the PM and the government do by way of policy decisions, proposing legislation and authorising expenditure is political in precisely the same sense.


  1. In a political constitution, calling an election is therefore, by definition, political. It is puzzling that critics appear to think that calling an election for political reasons in a political constitution is in some sense too political. It could even be argued that calling an election is one of the least controversial exercises of political power by the government because it is subjected to immediate direct democratic accountability at the hands of the electorate.


  1. This connects directly to the doctrine of confidence because that doctrine operates at two levels. 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.


  1. The current political situation is an unavoidable and powerful example of the negative impact of the FtPA on the doctrine of confidence and the balance of the constitution. The Benn Act was made an issue of confidence by the PM that led to rebels losing the whip. Very properly, the loss of those votes has spurred the PM to seek an election on two occasions, both rebuffed by the Commons. Seeking an election is the only alternative to resignation for a PM who has lost the confidence of the House.


  1. The resulting situation is an indictment of the FtPA. The PM has lost the confidence of the Commons, ostensibly, but cannot secure an election and there is no opposition figure around whom MPs can unite. This is a genuine impasse. It may be that the PM resigns and immediately brings a vote of no confidence against whoever is called to replace him (discussed further below). This kind of messy resolution is hardly satisfactory. Nor is it clear any longer what are the consequences, in terms of the doctrine of confidence, of losing a vote that has been made an issue of confidence by the PM.


  1. Worse still, the shadow executive that shepherded the Benn Act through the Commons refuses to bring down the government and subject itself to scrutiny and accountability at the despatch box, in relevant public bill committees, select committees and elsewhere. Instead, it hopes to make the government liable for the political consequences of a further extension to the Article 50 process by forcing it to act in a way that is flatly contradictory to its central policy. It would not have been possible to turn the government into a puppet in this way but for the FtPA.


  1. As an aside, a sharp contrast should be drawn between the Benn Act and ordinary Private Members Bills (‘PMBs’). As is well known, a successful PMB needs to have the cooperation of the Government in normal circumstances. Success in the PMB ballot by a backbench MP is frequently the prelude to some intense negotiations with those who control the business of the House to agree a mutually acceptable Bill that will get the nod in terms of the allocation of time. In a sense, the government effectively ‘adopts’ a PMB if it approves of it. The contrast with the machinations around Standing Order No. 14 and the passage of the Benn Act could not be starker.


  1. Many commentators, including this author, took the view that the two thirds requirement for a general election could not ultimately create an impasse such as this because the political optics of keeping a government in power would be so difficult for the opposition leadership. Current events have proven the contrary. The current state of paralysis has only been made possible as a direct result of the cul-de-sac created by the FtPA where confidence is apparently lost (though not in the narrow sense recognised by the FtPA) but an early general election is refused by the Commons. This has been exacerbated by the existence of an alternative, shadow, executive driving through legislation but refusing to install itself as the executive and accept the political consequences at the ballot box.


  1. In sum, the FtPA is antithetical to a long tradition of responsible government as it rebalances the constitution significantly in favour of the Commons at the expense of the executive that has its own legitimate democratic mandate through its manifesto put forward by an executive slate. The main justification for the Act appears to reside in an erroneous view that the political power to call an election is inappropriate in a political constitution. The current difficulties surrounding the Brexit process have been exacerbated by the FtPA in unhelpful ways.


What challenges arise for the political parties, the House of Commons and the civil service in the 14-day period following the passing of a motion of no confidence in the Government? Is there a risk of the monarch being drawn into the political debate during this period and, if so, how should this be mitigated?


  1. A successful vote of no confidence (‘VoNC’) is a rare but seismic political event. The changes made by the FtPA to add a 14 day period and a “vote of confidence” (‘VoC’) are an attempt to replace the previous discretion possessed by HMQ to take soundings and possibly replace the PM. On this point, this author strongly endorses the verbal evidence of Robert Hazell and Catherine Haddon which repeated the general understanding, promulgated by the former Cabinet Secretary Gus O’Donnell since 2010, that HMQ will no longer play any part in the delicate issue of appointing the PM. This new reality, it is suggested, is an inescapable necessity in a modern democratic polity. Instead, the elected politicians must sort it out, either through negotiations between the parties or, in extremis, by other methods.


  1. It is fair to concede that only some of the difficulties with the 14 day period are directly the result of the FtPA. Some difficulties and potential changes flow from the acceptance, set out in the last paragraph, that HMQ can no longer be involved under any circumstances. It is important to disaggregate these two different changes in assessing the new political reality.


  1. The starting point for understanding when and how a new PM can be appointed is the actual test which is that HMQ has a duty to appoint the person who is clearly best placed to command the confidence of the Commons. This test must be clearly distinguished from being able to command a majority of the Commons. They are not the same thing at all.


  1. If the latter were the test, there could never be minority governments even though these have long been an accepted feature of the constitutional history of the UK. Indeed, the appointment of David Cameron as PM, replacing Gordon Brown, was technically a recent example of the clearly best placed test because he was in fact appointed fractionally before the Coalition agreement was finalised by the parties. He had 306 seats. Brown had 258 seats. The timing was very close because a Coalition was formed shortly afterwards, but the details matter a great deal for our purposes. In sum, HMQ technically appointed a minority PM to replace a minority PM in 2010.


  1. As an aside, it is commonly thought that an ex-PM is supposed to ‘recommend’ his or her successor. This author respectfully dissents from this view. Of course, an ex-PM could not formally ‘Advise’ as to a successor precisely because an ex-PM has, by definition, no standing to give Advice at the point that someone else is best placed to command confidence. It is a matter of simple logic.


  1. The downgrading to a mere ‘recommendation’, it is suggested, seeks to gloss over the ‘Advice’ problem without genuinely confronting it. In reality, such a recommendation might be sought out of courtesy but is neither obligatory nor even necessary, particularly if the circumstances were somewhat fraught. It is also important to squash any suggestion that the incumbent has some kind of veto or delaying power. In truth, it is suggested that the recommendation as to who should be called to the Palace is now the responsibility of the House of Commons through the party system or, if all else fails, by other methods.


  1. It is tolerably clear that if a party or grouping has a majority then in the event a PM loses a VoNC, that grouping is the only possible source of a replacement PM. On one view, this approach explains how the current PM took office following a de facto loss of confidence (arguably) in the previous PM, even if that was never formally signalled by the Commons. If no candidate from within a majority grouping (where such exists) is forthcoming following a statutory VoNC, the PM is entitled to wait out the 14 days for an election to be triggered. The only way an alternative PM could emerge in these circumstances is if a faction or junior partner of a coalition broke away and joined the main opposition party.


  1. Much more difficult is where a minority administration loses a VoNC. This paper suggests that a simple test could help ensure that HMQ is never involved. The suggested test is that HMQ has a duty to dismiss, or accept the resignation of, a PM where an alternative candidate can demonstrate, (through for instance the number of signatures appended to a motion tabled to the House’s future business paper), the support of a greater number of MPs than the current whipped total of the governing party or grouping. This can be illustrated by considering the current Tory/DUP total of 297 whipped MPs. Applying the test in this paper would suggest that if another MP could demonstrate the support of 298 MPs, he or she would be entitled to be appointed as PM by HMQ.


  1. This test provides a neutral means of meeting the requirement that an alternative PM must be clearly best placed to command confidence because the issue is determined purely as a matter of numbers. For as long as an alternative PM has less numerical support than the majority grouping, it is never clear whether that MP has more support than a putative alternative candidate from the governing party or grouping. In those circumstances, the incumbent PM has no duty to resign because it is not clear if someone within the party, or someone outside the party, is best placed to command the confidence of the Commons in place of the incumbent.


  1. It is important to emphasise that the test proposed in this paper would be novel. Many take the view that the only circumstances where an alternative PM could have the right to insist on taking over is where they can demonstrate they could command an overall majority. This is reinforced, unfortunately, by the FtPA requirement that a new administration must win a positive VoC under the FtPA once appointed if an early election is to be avoided. It is suggested that this significantly reduces the chances of a viable minority government forming and also increases the chances of an incumbent PM clinging on in situations where, historically, they would have been replaced. The doubts over this question might have particular resonance in the next few weeks as the terms of the Benn Act become relevant.


  1. The lack of a clear and universally accepted numerical test, such as the one suggested in this paper, means that significant uncertainty now shrouds what could happen in the 14 day period following a VoNC under the FtPA. That uncertainty arguably increases the chance that HMQ could be drawn into the political arena. This is concerning. It is suggested that a clear and general numerical test provides a solid platform through which a non-discretionary set of norms could determine if and when a PM must resign in favour of an alternative candidate. Crucially, the test suggested here can accommodate the admittedly unusual circumstances where a minority government replaces a minority government. This argument is fleshed out in more detail in the posts listed in Footnote 4.


  1. The test suggested here, however, is in serious tension with the terms of the FtPA and in particular the requirement for a VoC which itself might encourage a reluctant PM to insist that any potential replacement PM first demonstrate they can command an overall majority despite not possessing an overall majority himself. This is not a criticism of the current PM who is entitled to work within the rules as they currently stand. It is suggested, however, that this difficulty, which could have a material effect on the Brexit process in October, is a further mark against the FtPA as it currently stands.


Would the prerogative power for the Prime Minister to dissolve Parliament and call a general election be revived in the event of repeal?


  1. This is a fairly complex legal question on which reasonable people disagree depending on how they conceptualise the nature of prerogative and its relationship with statute. I address the issue in depth at pp 492-499 of my Modern Law Review article cited above. My own view is that it would be possible to revive the prerogative power but that it would be better, instead, to confer a general power on the PM to advise HMQ to dissolve Parliament and provide for fresh elections to the House of Commons subject to a straightforward motion of approval in the Commons.


  1. At the risk of extending an already long submission, the short version of why it is possible, in my view, to revive the prerogative is that otherwise the 2011 Parliament will have successfully bound future parliaments. Unless a later parliament can completely unwind an earlier Act and restore the status quo ante, there is a sense that the ‘continuing’ theory of parliamentary sovereignty would be breached. Famously promulgated by HLA Hart, the continuing theory requires that later parliaments cannot be constrained by earlier parliaments and that must logically include a later parliament wishing to unwind entirely what an earlier parliament tried to do.


  1. Furthermore, it is important to mention that a bare repeal of the FtPA would not automatically revive the prerogative. This is because s 16 Interpretation Act 1978 specifically precludes the revival or pre-existing common law and other rules without express words to that effect which illustrate ‘a contrary intention’ by Parliament.


  1. If the committee is interested in a straightforward revival of the prerogative, the author commends the following draft Bill submitted by a parliamentary draftsperson (anonymously) on social media as follows:



  1. Revival of Royal prerogative power to dissolve Parliament for election.


(1) The Royal prerogative power of dissolving Parliament for the purpose of holding a parliamentary election, as that power existed before the coming into force of the Fixed Term Parliaments Act 2011, is revived and has effect.


(2) the Fixed Term Parliaments Act 2011 is repealed.


  1. A repealing Act would not be sufficient as drafted above. There would need to be further provision for the maximum term limits of four or five years (the author expresses no view on that issue), as well as other matters.


  1. Finally, the author recommends that if the FtPA is to be repealed and replaced, the best solution would be to replicate as far as possible the status quo ante but with the proviso that HMQ can no longer be directly involved because that would be inappropriate in a modern democracy. It is suggested that a power be conferred on the PM to ask HMQ to dissolve parliament as long as a simple motion to that effect had been passed by the Commons. The statutory power to dissolve would be technically held by HMQ, purely for reasons of tradition, sentiment and history, but the monarch would have no discretion to refuse under any circumstances. This is explained in more depth in the MLR article.


  1. One benefit of a simple motion to approve in the Commons would be to revive the previous rules on confidence which would arguably be unaffected by such a motion which would only be sought once the confidence question had been determined by the Commons. Secondly, if there was a genuine alternative candidate for PM, in accordance with the test set out earlier, such a motion could be refused by the Commons because the alternative PM would be able to muster more support than the incumbent against the idea of an election. The alternative PM could then table a motion to the effect that ‘This House believes [X] is best placed to command the confidence of the House of Commons’ or “That an humble Address be presented to Her Majesty, praying that she will be graciously pleased to appoint [X] to the position of Prime Minister”. If successful, the new PM would then take over if support for that candidate was thereby proven to exceed that of the incumbent.





  1. This paper has sought to address some of the ‘Key Questions’ posed by the Committee on the FtPA. It has argued that the FtPA has had a significant negative effect on the constitution for a variety of reasons. In particular, the effect on what remains a political constitution of a rigid, legalistic statute has been damaging and unhelpful at a number of levels. The common justification of the Act as reducing the power of the PM has been questioned both in theory and in practice and the paper has defended a conception of the political constitution that privileges a ‘responsible government’ approach to which the FtPA is deeply antithetical.


  1. The paper has further endorsed the view that HMQ now solely plays a ‘dignified’ rather than ‘efficient’ role in the constitution such that there are no circumstances where she could ever be involved in the selection of a PM. Instead, a novel test based purely on numerical superiority has been suggested for the selection of a PM in the House of Commons. Finally, it has been argued that the prerogative of dissolution could be revived but a better solution would be to confer on the PM a power to call an election but with the proviso that a motion in support of an election be approved in the House of Commons.


7 October 2019


[2] R (Miller) v Secretary of State for Exiting the European Union [2017] 2 W.L.R. 583, at [201].

[3] R (Privacy International) v Investigatory Powers Tribunal [2019] UKSC 22, at [39-40].


[5] A. Tomkins, Our Republican Constitution (Oxford and Portland, OR: Hart, 2005) 1.