Barely three months away from the election it is impossible to say who is likely to win: it could be either of the main parties, or it could be neither. Plenty of past elections have been too close to call but once the votes were in it was usually clear what had to happen next, even if in 1974 that meant cobbling together a government for a few months until it was time to have another go. The creation of five-year fixed-term parliaments – a rushed act of convenience with far-reaching and little considered constitutional implications – makes it much harder to know what may happen if there is no clear winner. Opinions about how easy it would be to engineer an election before the five years are up vary from no problem to no way (the truth is probably somewhere in between). What’s more, previous nail-biters were straight two-way contests. Even if no one emerged the winner everyone understood what winning meant: beating the other lot. This time round the permutations are almost limitless. We have become a multi-party electorate squeezed into a two-party electoral system.
Some of these permutations are not simply complicated but potentially incoherent. There are various scenarios that could make a mockery of the whole business. Suppose, for example, that the electoral arithmetic and the parliamentary arithmetic start to diverge. The Tories’ failure to get agreement about redrawing electoral boundaries means the national constituency map still favours Labour. It would be possible for Labour to win significantly more seats on fewer votes than its rivals: playing around with the numbers in the predictor grid on the website Electoral Calculus you can have Labour losing the popular vote but still ending up thirty or more seats ahead of the Tories. If that were to happen, Labour would almost certainly have lost the popular vote in England by an even larger margin, which would only add fuel to smouldering English resentment (Ukip would be there to fan the flames). Senior Labour figures have tried to counter charges of electoral bias by pointing to the million or more voters who are missing from the electoral register after a tightening up of the rules (the inference being that the people most likely to be missing – who include large numbers of students – are also the ones least likely to vote Tory). As the old saying goes, in politics if you are explaining you are losing. If you are stuck explaining how people who didn’t vote would have voted if they could have voted, you may have already lost.
The scenario in which Labour forms a government in Westminster despite an electoral drubbing in England rests on the assumption that Labour can continue to pile up votes and seats in Scotland. The continued rise of the SNP makes that less likely than it appeared even six months ago. But the fact that Labour might end up as the victim of the electoral system in Scotland – past a certain point the SNP would hoover up seats, leaving Labour with the scraps – shows how hard it is any longer to think of this as a general election. There is nothing general about it. National vote swings are a thing of the past. Different parts of the country will be having their own elections – in Scotland, in Northern Ireland, in Wales, perhaps in London, maybe even in Eastern England (Ukip-country) – where the vote swings and subsequent distribution of seats might be completely out of kilter with what is going on elsewhere.
If the SNP does return to Westminster with a raft of new MPs we face the possibility of the balance of power being held by a party that rejects Britain’s current constitutional settlement. The last time the SNP had sufficient representation in the Commons to make or break a government they used it to break one, calling the no confidence vote in the Callaghan administration in 1979 that helped usher Thatcher into Downing Street. Similar difficulties would inevitably arise if Ukip secured more than a handful of MPs, which is a lot less likely but by no means impossible. What happens then? Can you cobble together a government with people whose primary purpose is to change the terms on which you govern? Neither of the two main parties would be well advised to try that with Farage, who will be agitating from the off for anything that might unravel the relationship with Europe. The idea of binding Ukip into any government for five years seems fantastical, which means the parliament would effectively run until a EU referendum was held, at which point anything could happen.
In the case of the SNP the problem is a mismatch between the party’s ideological commitments and its strategic interests. As a party of the left it can hardly do a deal with the Tories. But as a party of independence it can hardly do a deal with Labour, given that Ed Miliband needs the UK to stay together if he is going to be prime minister of anywhere. It doesn’t help that Tory irrelevance in Scotland leaves Labour and the SNP free to take out their grievances on each other. One of the deep ironies of recent British politics is that the SNP has seen its prospects best served by success for David Cameron. If the Tories do hold a referendum on Europe in 2017, and England votes to take Scotland out with it, another referendum on Scottish independence would be hard to resist. It might take a Machiavellian political genius to negotiate a way through those perilous waters. Then again, Alex Salmond hopes to spend the next five years doing his politics in Westminster, so go figure.
If coalition government proves impossible, then we are looking at a minority government, which would allow deals to be done and undone on a more ad hoc basis. Keeping that up for five years appears pretty unlikely, but even to get it started, one of the two main parties would need to have a claim to go first. What happens if the election is a dead heat? At the moment the chances of the two ending up with almost the same number of seats, though remote, are certainly no more remote than any other possible outcome. A tie might produce a stalemate. Who breaks the deadlock? Not the queen, please god. Who else then?
It is conventional for politicians to say that they are not interested in opinion polls because the only poll that counts is the one that happens at the ballot box. But if the ballot box gives a confused and confusing answer, why shouldn’t the public continue to be polled on what it wants to happen next? It’s not as if anyone will have had the chance to vote for any of the possible permutations – none of those is on the ballot paper – so it seems reasonable to canvass opinion on the emerging options. And even if it were unreasonable, in an age of instant online polling it’s going to happen anyway. In the event of deadlock, public opinion will have to be factored in. Yet the difficulty of knowing how to read the current state of public opinion – beyond the fact that a lot of people seem to be pissed off about a lot of different things – is what makes this election so unpredictable. It’s not going to make forming a government any easier.
It is tempting to laugh at the 535 days Belgium went without a government during 2010-11, following a federal election that produced an effective dead heat from a deeply divided electorate (part of the joke is that the Belgian economy outperformed its European rivals over the same period). But we are not Belgium (at least not yet). The selling point of the Westminster model has always been its decisiveness: you might not like the result, you might not even think it fair, but at least it’s quick. The removal vans for one lot should be round the back before the other lot have their keys in the front door. In 2010 it took five days to form a government and a general sense that it was taking too long was in the air sometime around the afternoon of day four. Here is the final and deepest uncertainty surrounding this election. How long will people be willing to put up with the mess it might produce? Will they care enough to demand a change to the system that produced it? The public’s tolerance for political confusion is said to be pretty limited – and we are always told that the tolerance of the markets is more limited still – but it may be that an extended period of confusion is what they get. If people are fed up with politics now, how much more fed up are they going to be if this election only makes things worse?
On the other hand, it is usually a mistake to anticipate a crunch point for the British electoral system. In 2004 I wrote in this paper (21 October) that a crisis of legitimacy would follow if Tony Blair were returned to power with a big majority on barely a third of the votes of not much more than half the electorate. That is what happened: he won a majority of 66 on a vote share of 35.2 per cent from a turnout of 61.4 per cent, meaning that his government retained the enormous powers afforded to the victors by the British parliamentary system despite the fact that if you gathered 100 eligible voters in a room only 22 of them would have voted Labour. Did the other 78 repudiate the result? Hardly. They just shrugged and got on with the low-level grumbling that is the default condition of most modern electorates. It could be argued that the present levels of mistrust in mainstream British politics have their roots in 2005: from Blair to Brown to the coalition via one single further election has provided plenty of space for the grumbling to turn into outright contempt. Still, the British system has been through worse in the past and survived more or less intact.
Genuine crises of legitimacy are rare. This is not at present an especially well-governed country – it is not hard to imagine how it could be done better, and it could certainly be done more fairly – but by any historic standards the UK remains a prosperous and peaceful nation. The unambiguous markers of political failure – widespread civil unrest, economic collapse, armed intervention – are not on the horizon. We are not Greece. We are not even Spain. Not yet. That is part of the problem. How long can a functioning democracy continue to function with a political system that is broken? Who knows. But it could be quite a while.