Uncertainty is a constant and deleterious feature of the Brexit experience. The prospect that the United Kingdom (UK) might – or might not – soon be leaving the European Union (EU), on terms and at a date that are uncertain, is a source of immense destabilisation. It has many manifestations, for instance in the realm of economics and trade. One area in which Brexit has both exposed and augmented ambiguity involves the rules according to which the political system itself functions: the constitution. It has created doubts about the place of the different components of the UK within the Union and the position of the devolved institutions. It has raised uncertainties about the future operation of the legal system; and about the relationship between the UK executive and the Westminster Parliament. The latest instalment in this prolonged period of systemic ambiguity and instability pertains to the tenure of prime ministers.

The legal position regarding the appointment (and dismissal) of premiers is clear. The monarch possesses and exercises this power personally under an ancient set of residual regal authorities known as the Royal Prerogative. But, as is often the case with the UK constitution, particularly where the role of the monarch is concerned, the practical reality is different to and more complex than the formal arrangement. Some very powerful and deep-rooted principles apply to the deployment of this power. They are fundamental to the status of the UK as a parliamentary democracy with a limited, constitutional monarchy.

The monarch, so the received wisdom runs, should not be seen to be involved in matters of party-political controversy or to have taken sides in partisan contests. From the point of view of the ruler, to allow such a perception to develop would be to undermine the monarchy itself. Perhaps more importantly (though it is rarely phrased in these terms out of deference to royalty) it would also be problematic from a democratic perspective were the head of state overtly to interfere in politics.

The appointment (and dismissal) of prime ministers is one of the most obvious areas in which restraint is required to avoid such an outcome. Normally, it is relatively easy for a monarch to maintain neutrality in this regard. The guiding principle is that they must appoint as Prime Minister the person best able to command the confidence of the House of Commons. If a single party holds a majority in the House of Commons, then whoever that party has chosen as its leader should be the Prime Minister. That person might come to the role from opposition having won a General Election (the most recent person to reach No.10 in this way was Tony Blair, Labour leader in May 1997); or because they attain the leadership of a party already in possession of a Commons majority (as, most recently, did Theresa May in July 2016). In such circumstances, there are no doubts, constitutionally, about what is supposed to happen. All the key players involved accept the outcome (including the outgoing Prime Minister, who offers their resignation, so there is no need to dismiss them). The monarch does not need to make a choice about who should govern.

If no single party has a majority in the Commons, matters can become more complicated. An inconclusive General Election can produce doubts about who will be able to govern. In this circumstance, the sitting Prime Minister is, by convention, allowed to continue in post and test whether they have the confidence of the new Commons. They may, however, choose to resign if it becomes apparent that a potential successor has, through political negotiations, constructed a bloc of MPs, comprising two or more parties, that consents to the formation of a replacement government. A further understanding that some believe to apply is that an incumbent Prime Minister should not leave prematurely; and should await the clear conclusion of an agreement between other parties before exiting their post. This proposition, for those who accept it to be the case, implies a duty to ensure stability and continuity in government.

In recent times, we have seen two general elections that failed to yield an outright victor. In 2010, the Labour Party lost its majority, and its leader Gordon Brown, was succeeded by a coalition led by David Cameron of the Conservatives, with support from the Liberal Democrats. In 2017, the Conservatives, under Theresa May, lost their majority, but May was able to stay on at the head of a minority, single-party government, having secured backing in the Commons from the Democratic Unionist Party. In both cases, there were ultimately no serious constitutional doubts about the process; and political negotiations resolved the matter. Active involvement from the monarch was unnecessary.

However, we should not assume that such processes need always run as smoothly. Currently, the Conservative Party is holding a leadership contest, triggered by Theresa May, the outgoing leader. If the entire Conservative cohort in the Commons, along with Democratic Unionist MPs, is apparently supportive of their being able to form a government, then there is no doubt that the Queen should appoint the winner as Prime Minister. But what if that support is not forthcoming?

At the time of writing, Boris Johnson remains the most likely victor. It is well known that a significant number of MPs in his own party do not merely regard him as a less than ideal candidate. They hold pronounced doubts about his very suitability for high office; and are vehemently opposed to his stated determination that the UK should leave the EU at the end of October, even if no exit deal is in place. Some MPs seem willing to offer resistance to Johnson and this course of action that go beyond simple rebelliousness. For instance, they might be willing to take part in attempts to deny financial grants to the government; or vote for a ‘no-confidence’ motion in the government, perhaps within the terms of the Fixed-term Parliaments Act 2011, potentially leading to an early General Election, or perhaps outside the terms of this Act. Given the present parliamentary arithmetic, and that the Conservatives even at full strength lack a majority of their own in the Commons, the position is precarious.

It is possible that, in advance of the conclusion of the Conservative leadership contest, that there will be serious doubts about the ability of the winner (presumably Johnson) to command the confidence (however defined) of the Commons. It may have become apparent by this point that a substantial number of Conservative MPs – exceeding in size the Commons majority that the cooperation of the DUP provides – are so opposed to him personally and for his stated policy that there are good grounds to believe his position as Prime Minister would be untenable. What, precisely, ‘untenable’ means is of course a matter of judgement. The May premiership arguably seemed untenable from the point at which she lost the first vote on the exit deal on 15 January 2019, but she did not announce her intention to leave until 4 May, and remains in post at the time of writing. But when the Queen first appointed May as Prime Minister, her position seemed secure. Bringing someone into the post who already seems unable effectively to govern would be a different matter.

The monarch could be in an uncomfortable position. Precedent is a vital commodity in times of constitutional doubt. Previous decisions can provide guidance, or at least be used as a source of legitimacy for a given course of action. But analogous circumstances that might be source of useful precedent, in even relatively recent times, are lacking. When James Callaghan was elected Labour leader in succession to the sitting Prime Minister, Harold Wilson, in 1976, Labour had a small Commons majority and there was no open revolt against Callaghan among his parliamentary cohort. The Conservative changeovers from Anthony Eden to Harold Macmillan in 1957 and Harold Macmillan to Alec Douglas Home in 1963 placed the Queen in an awkward position. But the Conservative majority was substantial, and there was no question that Macmillan and Home could command the confidence of the Commons. The problem on these occasions was that the Conservative Party lacked a transparent method of choosing leaders. The means by which Rab Butler had been passed over on both occasions was obscure, causing resentment among his supporters. The Conservatives then adopted a leadership election mechanism to avoid a further repetition of this circumstance. But having a clear means of choosing a leader will not necessarily resolve matters in 2019. The Conservative Party might have chosen a new leader in accordance with its own rules, but the rules of the UK constitution are that a government needs to be able to command the confidence of the Commons. If there are serious doubts about the new leader of the Conservative Party being able to achieve this support from the Commons, a series of questions arise. Guidance from precedent is hard to find, so the monarch and her advisers might have to exercise their own discretion in answering them. They engage a matter of exceptional political controversy, and whatever decisions the Queen implements will risk embroilment within it.

A first question involves whether the monarch is obliged to appoint the new Conservative leader as Prime Minister. Some – including but not only partisans of the would-be premier – will argue that she is. But this claim, like many other aspects of the UK system, is difficult definitively to establish, and certainly has no legal basis. To reverse the premise, it might be asked, is the monarch required not to appoint someone who appears to lack the confidence of the Commons? But if so, how should the basic requirement of a functioning government be fulfilled? Perhaps, for the time being, Theresa May could be induced to remain in post. Indeed, one interpretation of the relevant principles discussed above suggests that May has a duty to do so, if a stable successor administration is not apparent (though there is a widely accepted view that the Prime Minister cannot advise on the succession with the same force that they can on other deployments of the Royal Prerogative, and that the Queen can take advice on this question from whichever sources she as monarch deems appropriate). But this option would not be a long-term solution. May would appear very much a caretaker. It would be difficult – and probably constitutionally inappropriate – for her government to take any major decisions that might bind a successor administration, unless on a basis of inter- and intra-party consensus, which would be difficult to establish. Yet the current Brexit environment means that a government able to take major decisions is needed at least as urgently as at any point in the peacetime history of the UK. Perhaps May could remain in post in what was effectively an acting role for a limited time while Johnson, given a conditional mandate by the monarch, attempted to secure necessary assurances from within his parliamentary party. (This practice, which was employed in 1963 before the appointment of Home, was also used in earlier eras in such times of uncertainty, and described by Anthony Trollope in his novel The Prime Minister, in which an earlier Queen commissions the fictitious Duke of Omnium to try to establish a Liberal-Conservative coalition government.)

Assuming that Johnson by some means becomes Prime Minister, the potential for complexities surrounding the use of royal authority will not necessarily cease. He might, for instance, ask the monarch to prorogue Parliament to prevent it from interfering with his policies. Such a request would once again place the Queen in a position where controversy was hard to evade. In the divisive atmosphere of Brexit, it is difficult to avoid taking one side and infuriating the other. A scenario in which neutrality was impossible might also arise if the House of Commons and House of Lords passed a Bill that sought to restrict or impose a course of action upon the government with respect to Brexit, and in response the Prime Minister asked the monarch to withhold Royal Assent from it.

There is also a chance that the royal power to appoint and remove the Prime Minister would become a live issue not only before, but relatively soon after the formation of a Johnson government. As I have discussed elsewhere (Brexit and Parliamentary “sovereignty” and Parliament: sovereign or supine?), a Parliament that wants to prevent a no-deal Brexit, or delay Brexit, or bring about another referendum, or revoke Article 50, or achieve some combination of the aforementioned goals, faces an important obstacle. However creative it may be with procedure, legislation and financial measures, its chances of success are limited while the occupant of No.10 is determined to follow a different course. To maximise its prospects, a Parliament with these intents must remove the Prime Minister and replace him with a different premier at the head of a different government. Leaving aside the immense rupture in the party-political system that it would involve (that I and other authors on the Federal Trust website have previously addressed, see for example The Brexit Revolution Eats Its Conservative Parents and other Blogs), I will concentrate now on the constitutional aspect of such a manoeuvre.

The initiative must come from the Commons. Convention – supplemented, but also made more opaque and perhaps weakened by the Fixed-term Parliaments Act 2011 – dictates that a government remains in office subject to possessing the ‘confidence’ of the House of Commons. There has never been a precise, agreed, and comprehensive definition of this concept. We know that it does not mean that a government needs to win every vote to remain in office. Beyond that, confidence has been defined partly subjectively, on a basis of the circumstances of the time. The clearest way that the Commons can deny confidence is through voting for a motion expressly to this effect. Following the 2011 Act, there seem to be two basic options open to the Commons if it wishes to oust Johnson. First it could vote for a no confidence motion that used the wording set out in the Act. A 14-day period would commence. If, by the end of it, the Commons had not voted for a positive confidence motion in the existing government or a different government (again using phrasing prescribed in the Act), a General Election would take place.

How effective a General Election might be at resolving present difficulties is an important subject in its own right. But for present purposes, let us assume that the purpose of a no-confidence motion under the Act would be to replace the present Prime Minister and vote confidence in a new administration. There is no clear guidance in the 2011 Act – or anywhere else – as to proper procedure for affecting this outcome during the 14-day period. Moreover, since this procedure mandated under the Act has never been instigated we do not even have precedent to invoke. A sitting Prime Minister might wish to remain in office, awaiting a General Election, arguing that the people should decide. Even if there was clear evidence that a Commons majority had assembled around an alternative Prime Minister and government, the monarch would, yet again, face a difficult decision. Would she be willing to dismiss a premier she had appointed so recently, and appoint a successor who would probably not be the recognised leader of a parliamentary party?

If the Commons wished to keep the issue of supplanting the Prime Minister separate from that of an early General Election, it might vote for a motion that clearly expressed a lack of confidence in the government, but did not use the exact wording contained in the 2011 Act. For this approach to work, the Commons majority would need to have a replacement government ready, and probably to have communicated the fact of its existence to relevant players, including the Palace and presumably the general public, in advance. The Prime Minister might deny the constitutionality of this approach, claiming that confidence motions outside the terms of the Act 2011 were no longer valid. He would probably have strong backing from Eurosceptic forces, complaining of an elite conspiracy to subvert democracy. A difficult choice could once again fall to the Queen and whoever was advising her.

When confronted with one of the treacherous scenarios set out above, the Queen might seek to avoid making a difficult choice herself by encouraging the holding of cross-party talks to resolve the matter. The last time a monarch performed such a role was in 1931, when George V facilitated the formation of the National government. However, in the intervening period, expectations surrounding the role of the monarch have changed. Such action would be subject to far closer, and critical, scrutiny today. That the Queen had appeared to advocate a coalition would itself be controversial. And that government, were it formed, would need to adopt a policy of some kind towards Brexit, with which the monarch would by extension be associated. A further important distinction from the 1931 position is that an intervention by the monarch in 2019 could entail the removal of a sitting Prime Minister. In 1931, George V was helping Ramsay MacDonald to continue in office, at the head of a newly formed coalition.

This discussion reveals two major problems facing the UK. One is a long-term, structural issue. Under whatever democratic system is used in any country – with a directly elected executive or a government resting only in Parliament – at times, the head of state or someone acting on their behalf is likely to be needed to resolve constitutional deadlocks. The UK monarchy lacks the democratic legitimacy to do so effectively. Whether and how we will resolve this issue is a matter for the future. But there is a more pressing matter that confronts the UK at this juncture. It faces a default position of exit from the EU without a deal at the end of October, and a Prime Minister who presents this outcome as viable, and even desirable. Those in the Commons and beyond who take a different view must prepare for the constitutional battle that awaits.

Further reading:

Comment on “Brexit – By Royal Appointment?”
by Professor Sam Whimster, 22nd July 2019