Being able to carry the support of Parliament is the key determinant of being a government. The issue is what kind of demonstration or defeat signifies that a government has failed to command confidence. Since 1782, the most clear-cut way in which this occurs is when a government is defeated on a vote of confidence – either a motion tabled with such wording (either referring to lack of confidence or a censure of the government) or on other votes that the government has indicated it considered a matter of confidence.
FTPA challenges the previous convention purely because it explicitly sets out a process for what happens if a particularly worded vote of confidence occurs. What is less clear is how other worded votes of confidence are used by Parliament and whether past convention requiring a government to resign remains in force. We won’t entirely know unless the FTPA process is actually used or deliberately avoided. History suggests that convention will continue to be as powerful and current parties may just avoid triggering the FTPA.
FTPA is not about forcing a government to resign, but rather a mechanism to allow for an early election. A defeat on an explicitly worded vote of confidence, and the failure to achieve a second vote of confidence within 14 days, is one of only two ways in which an early election can be called.
Confusion exists about what effect this has on other worded votes of confidence or critical votes such as the Queen’s Speech or a budget. Certainly they don’t trigger the 14-day period – the Act is clear on this – but ought a government resign if it loses any of them? In practical terms this is about whether a government returned in May should resign if it fails to carry a Queen’s Speech, or insist that without a formal FTPA worded vote of confidence that it will continue. Or, because the Act does not require resignation, could a government defeated on one FTPA vote of confidence stay on through the second vote of confidence and get a second election instead of passing power to its opponents? Both questions are matters that are more about what will look legitimate, fair and politically viable, rather than a question of legal interpretation. They go to the heart of past convention of when and why governments resigned when they lost.
To understand these questions, it is best to look at the history of votes of confidence, what form they took, and how governments responded to them. This not only helps understand the context and meaning of the FTPA and its particular form of confidence vote, but also to consider why the previous convention still carries weight and how much this might guide parliamentary practice.
Retaining the confidence of the House of Commons is a core principle of the UK constitution. In 1782, a motion censuring the government of Lord North for its conduct of the war with America was the first time a government was brought down by a vote of confidence. Since then, there have been 20 government defeats on a vote of confidence all leading to either dissolution or resignation (and if the latter, a second government being formed in their stead).
Prior to FTPA, following the 1782 precedent, governments were expected, or required, to either resign or seek a dissolution if they lost a vote of confidence. Whether a vote was considered a matter of confidence was something that was signalled to the House before the vote by the government. Some motions explicitly mentioned confidence such as the March 1979 vote that brought down the Callaghan Labour government; others were votes that were declared a matter of confidence by the government itself, like the October 1924 amendment to a censure motion that brought down the minority Labour government.
Other votes are considered so important that governments have been expected to treat them as seriously as a confidence vote, but do not necessarily require resignation or dissolution in the same way. As a House of Commons Library note on this topic sets out, the suggestion that a Queen’s Speech or budget is tantamount to no confidence is not clear.
A Queen’s Speech is the first indication that the government can demonstrate it can command confidence. However, the vote in itself is not a strict confidence motion. In 1924 the Conservative government was defeated on its King’s Speech, but through an amendment that added the words ‘but it is our duty respectfully to submit to your Majesty that Your Majesty's present advisers have not the confidence of this House’. In fact, of the five Queen’s or King’s speech defeats in our history, all of them were defeats on amendments to the motion and explicitly mentioned confidence. So the question of whether it is akin to a confidence motion is rather moot – oppositions wishing to defeat a government at this early stage used a confidence motion to do so. More significantly, usual custom had been for governments to meet Parliament after an election, be defeated and then resign. Disraeli in 1868 was the first Prime Minister who resigned before Parliament met, indicating that the election result, not the will of the House, was the key determinant. The budget is also one of the key votes for any government, as the government’s dependence on the House for supply (finance) is a core principle in our constitution. However governments have been defeated on their budget only twice – both in the late 19th century. In each case the government resigned, to be replaced by the main opposition, but the motions were not worded as a vote of confidence. So this is a political convention – both governments recognised the seriousness of the defeat and decided to resign. The precedent for today is less clear.
Another issue is how governments react to a defeat. Governments fall or prime ministers resign for a number of reasons, not just because of a defeat in the Commons. Of those 20 confidence defeats, only four have resulted in the government calling for a dissolution of Parliament and a new election. The remaining 16 all saw the government resign and a new one take its place (in some cases this successor government calling for an election, in other cases continuing to govern for a period before asking for a dissolution). This is relevant because a change from one party to another in government without an election may seem alien, but was not uncommon if you go back far enough. The question now is whether, with the political culture we have today, changes of government further into the next parliament (say in two years’ time) will seem legitimate to the public without another election.
All of this could be constitutional dancing on the head of a pin. The political reality would probably be much simpler – a government defeated on its Queen’s Speech will be under great pressure to resign and would risk its reputation if they used the letter of the law to stay in office. But our history again provides useful reminders of where politicians have at times pushed back against custom or used stricter interpretation of the rules to stay the course.
In December 1923, Conservative Prime Minister Stanley Baldwin decided to continue in office as a minority government, despite clear indications from other parties that he would not have sufficient support in his King’s Speech – on which he was then defeated in January 1924. He did so partly because the Conservatives wanted to demonstrate that the Liberals, whose political fortune was on the wane, could only gain power with the support of Labour. And in March 1974, with a minority of minus 17, the Labour Prime Minister Harold Wilson let it be known that only explicit votes of confidence would be enough to force his government to resign. Labour survived as a minority until they called a second election in October, and then managed a majority, but only of three and that was then whittled away, before the Government was eventually defeated by one on a vote of confidence in 1979.
Whether all of this becomes an issue depends on the result and on the type of government formed. Even if FTPA is avoided, amended or repealed, under minority government the way in which confidence votes are treated will come down to political realities and perceptions of legitimacy.