When the Numbers Fail: A Statutory Code for Confidence and Dissolution

The Dissolution and Calling of Parliament Act 2022 was a sensible correction. It repealed the Fixed-term Parliaments Act 2011, restored the prerogative power to dissolve Parliament, and removed the statutory fiction that a government could be simultaneously defeated and immovable. Few constitutional reforms of the past decade were as clearly warranted. Yet repeal answered only the question of legal power. It did not answer the political questions that now sit at the centre of Westminster: what counts as a confidence vote, who certifies the numbers, and what advice a Prime Minister tenders to the King when the Commons will not resolve itself.

That is the confidence gap. It is not a crisis. It is a small, corrosive ambiguity in the operating manual of the British constitution, and it is the kind of ambiguity that tends to be exposed at the worst possible moment.

The repeal was right; the silence was not

The case against the Fixed-term Parliaments Act was always stronger than the case for it. The Act created a statutory language of confidence that did not match the realities of Westminster. It allowed a government to lose major votes without losing office, encouraged brinkmanship, and left the country in 2019 with a Parliament that could neither govern nor dissolve itself. The 2022 Act restored the older, more flexible arrangement: the Prime Minister advises, the King acts, and the Commons retains the ultimate political sanction.

But flexibility is not the same as clarity. The 2022 Act did not define confidence. It did not set a procedure for a hung Parliament. It did not say how the Speaker should record a confidence defeat, or what should happen if a Prime Minister who has lost the House refuses to resign. Those questions were left to convention, and convention is at its weakest when the arithmetic is at its most fragmented.

Confidence has become a word without a procedure

Not every government defeat is a confidence defeat. In 2019, the Commons defeated the government on Brexit motions, on the withdrawal agreement, and on prorogation, but none of those divisions was treated as a formal confidence question. In 1979, by contrast, James Callaghan's government fell by one vote on a motion of confidence. In 2017, Theresa May's government survived through a confidence-and-supply agreement with the Democratic Unionist Party. The difference between those cases was not legal doctrine. It was political judgment, exercised in the moment.

That is not a satisfactory basis for a modern constitution. Standing Orders provide no settled definition of a confidence motion. The Speaker's authority to select amendments is not a power to certify confidence. A Prime Minister can govern after losing significant votes, while an Opposition can claim a mandate to form a government without ever testing the House. The result is a system in which everyone claims to know what confidence means until the moment it actually matters.

The Crown's position is convention all the way down

The monarchy's role is often described as ceremonial. In a hung Parliament, it is anything but. The King acts on ministerial advice, but the conventions governing that advice are not codified. The Cabinet Manual describes the caretaker convention and the expectation that a Prime Minister who cannot command the House will resign or seek a dissolution. The Lascelles Principles, formulated in 1950, suggest circumstances in which the monarch might refuse a dissolution: if the existing Parliament is still viable, if an election would be harmful to the economy, and if another government could be carried on. But the principles are informal, contested and not binding.

In practice, the Crown is asked to adjudicate a political arithmetic it cannot itself verify. If a Prime Minister loses a confidence vote and refuses to resign, or if a hung Parliament refuses to resolve, the palace becomes the referee of last resort. That is not a dignified position for a constitutional monarchy. It is also not fair to the King, who should not be drawn into partisan calculations about whether the numbers might change next week.

Four clauses, not a written constitution

The answer is not to resurrect the Fixed-term Parliaments Act under another name. It is a short Dissolution and Confidence Act, with four narrow provisions.

First, define a confidence motion. A motion should count as a confidence motion only if it is designated as such on the Order Paper by the Leader of the House or the Official Opposition and selected by the Speaker. That preserves the political nature of the question while giving it a formal label.

Second, require the Speaker to certify the outcome. This is a clerical act, not a political one. The Speaker would certify the result of a designated confidence division, not pronounce on who should govern. The certification would be published and laid before the House.

Third, set a short clock. If a government loses a designated confidence motion, the Prime Minister should have fourteen days to demonstrate that an alternative government can command the House, or to advise dissolution. If neither happens, dissolution should follow by default. The objection to the Fixed-term Parliaments Act was that it created zombie government. A default-dissolution rule answers that objection: the Commons cannot indefinitely refuse to choose.

Fourth, protect the Crown. The Act should confirm that the monarch's reserve powers remain intact and that ministerial advice must be consistent with the statutory code. It should also make clear that the courts have no role in adjudicating confidence, save for ordinary procedural review. The point is to keep the Crown out of the arithmetic, not to invite judges into it.

The conservative case for clarity

Conservatives should be the last to worship codification. The strength of the British constitution is its capacity to bend without breaking, and a written constitution would be a constitutional revolution in all but name. But a short statutory code is not a written constitution. It is a procedural settlement, limited in scope and conservative in temperament. It preserves the prerogative, respects parliamentary sovereignty, and protects the monarchy from being dragged into disputes it cannot resolve.

The confidence gap is not the most urgent issue facing the country. But it is the kind of constitutional loose thread that, if left unpulled, can unravel at the worst possible moment. A short Act in this Parliament could tie it off without pretending to solve every question of governance. That is not codification for its own sake. It is maintenance of the machinery of government, and it is overdue.