Were it not for the coalition anxieties of the Conservatives and Liberal Democrats, the 2017 parliament would have almost certainly have come crashing down by now.
There is arguably no dependable working majority for any policy Theresa May wishes to pursue, let alone the one her government exists to enact: Brexit. Burdened with temperamental coalition partners in the shape of the DUP, her sclerotic executive has effectively given up the pretence of acting as a normal government would.
It accepted swathes of Labour and SNP amendments to the Budget – which the DUP threatened to vote down over Brexit – in order to avoid defeat. It has also pursued a policy of deliberate disengagement when it comes to opposition day debates – a tacit acknowledgement of its inability to reliably command the confidence of the Commons.
That fundamental weakness will be thrown into harsh relief when MPs vote down the Withdrawal Agreement next week. Convention dictates that the rejection of legislation of such importance would effectively amount to a vote of no confidence in the government. Cue the dissolution of parliament and a general election.
But of course, the introduction of 2011’s Fixed Terms Parliament Act – designed as a sop to those who believed the Conservatives could pull the rug from under the Coalition for political gain – circumvents that convention. The means by which a general election can happen have been severely limited as a result, and zombie governments like May’s now have a life support machine on the statute book.
The letter of the law means that a crushing defeat on anything that isn’t a motion of confidence in the government can be brazened out by a government without scruples, regardless of how essential it is to its ability to govern (the Budget, for example, or the Withdrawal Agreement). Instead, there are two ways that Theresa May could go to the country – calling for a general election herself, or blundering into one.
Although the Fixed Term Parliaments Act turned the ability to dissolve parliamentary from a prerogative power wielded at the prime minister’s discretion to a statutory one that is in theory heavily circumscribed, May could still call an election whenever she wanted. The provisions of the Act allow for an early dissolution, provided two thirds of MPs vote for it, as they did in 2017. It requires opposition MPs to acquiesce, and there is always a political imperative for them to do so. For obvious reasons, however, May is not going to do so before 29 March 2019.
The only other route to an early election – and the only plausible one in the current climate – is the defeat of the government in two separate votes of no confidence, which can be tabled by any MP but in practice must come from the leader of the opposition, whose motions are guaranteed parliamentary time. Should the government lose the first, a two-week “cooling off” period follows, after which there is another vote. Should it fail to command the confidence of the house for a second time, or no alternative government emerge that is able to do so, then parliament is dissolved and an election ensues.
The arithmetic of the 2017 parliament means that any such vote would be incredibly tight indeed. But perversely, it is only likely to happen if the government succeeds in passing its flagship policy in its current form. The DUP has said it will only vote against the government in a confidence motion should the Withdrawal Agreement – and with it the Northern Irish backstop – pass the Commons. The Labour leadership has said it will not table a motion of confidence until such time as it is likely to win it, which would at the very least require the DUP to go nuclear and vote against the government.
We know the DUP will not do this if the Withdrawal Agreement falls, as it is still overwhelmingly likely to do. And with that the political avenues to such a vote are further constrained. There are other reasons the DUP might not prioritise an election too: its Westminster leader, Nigel Dodds, is defending a slender majority in North Belfast, and even if it repeated or bettered its historic haul of 10 seats from 2017, results in England, Wales and Scotland would risk the loss of its kingmaker status. The government’s position is further bolstered by Sylvia Hermon, the independent unionist member for North Down (and only other Northern Irish MP to take her seat). Hermon has said that she could never countenance putting Jeremy Corbyn into Downing Street.
If it can count on the support of those 11 MPs from Northern Ireland, the government’s job of winning a majority of 635 MPs – the number you get if you subtract the seven abstentionist Sinn Fein MPs, the Speaker and his three deputies and four tellers for any vote from the 650 total (two Tory, two Labour) – is much easier. The Conservatives have 316 MPs minus non-voting deputy speaker Eleanor Laing, which, minus two tellers, takes them within four MPs of victory. Hermon’s support would take them to 316, one clear of safety. To compete, Labour would need the votes of every other opposition MP in addition to its own 255 (257 minus deputy speakers). Minus two Labour tellers, and with the SNP (35), Liberal Democrats (11), Plaid Cymru (4), former Lib Dem MP Stephen Lloyd (1), suspended Labour MP Kelvin Hopkins and Jared O’Mara, Frank Field, John Woodcock and Ivan Lewis, the former Labour MPs who left the party in 2018 (5), they total 309.
The road to success or failure is thus the same as it has been throughout this parliament – winning the support of the DUP’s 10 MPs, who would take either camp clear of the 318 threshold. The Labour leadership’s reluctance to call a confidence vote until such time as it has withdrawn its support from the government is testament to this. Until its position changes, an election via the no confidence route is unlikely. And while some Conservative MPs are willing to entertain voting against their own government in the first of the two confidence votes in order to kill off a no-deal exit, it is unclear how they would vote in the second – and whether their preference would be for some cross-party solution rather than an election.