If the latest opinion polls play out in the general election, John Curtice argues, the ensuing fracas of post-election negotiations and the prospect of a minority government are likely to provoke serious debate about Britain's electoral system and wider constitutional settlement.

The truth is none of us can be sure what the result of the election on 7 May will be. Not since 1992 has the gap between the Conservatives and Labour in pre-election polls been so narrow – and on that occasion the polls proved to be wrong. Otherwise, we probably have to go back to the unexpected election of February 1974 to find the last contest that looked remotely as close as this year's promises to be. To complicate matters, Ukip, the Greens and the SNP in Scotland are all recording unprecedented levels of support, leaving us wondering whether this poll support will be realised in the ballot box and, if so, with what effect.

But if there is a lesson to be learnt from February 1974 it is that any outcome remotely resembling that suggested by recent opinion polls is likely to reopen the debate about electoral and constitutional reform. Forty years ago no party won an overall majority and the party that won the most votes did not secure the most seats. Third parties – primarily the Liberal Democrats and the SNP – won an unprecedented share of the vote, yet in the Liberal Democrats' case this brought very little reward in terms of seats (on this score, the SNP, however, did relatively better). But if the first-pastthe- post system was doing the job it is expected to do, somebody should have won a majority, that somebody should have been the party with the largest share of the vote, and nothing like a fifth of voters should have been voting for the Liberal Democrats or other smaller parties. In some eyes at least the system was seen to have 'failed', and its continued use became the subject of considerable discussion and scrutiny.[1]

Three months out, the 2015 election appears to be holding out the possibility that some and maybe all of these features of the February 1974 result will be repeated. There seems to be a significant chance that no single party will secure an overall majority. There is at least some risk that the party with the most votes will not necessarily win the most seats. Meanwhile, at least two of the 'insurgent' parties, Ukip and the Greens, may secure very little by the way of representation in the House of Commons, while a third, the SNP, seems likely to fare rather better and might even secure more than its proportionate share of seats. Any outcome that comes close to resembling this picture would raise important questions about how Britain conducts its elections.

However, there is one vital difference between what happened in February 1974 and the prospects for May 2015. On that occasion, Labour's leader Harold Wilson, armed with just four more seats than his Conservative rival Edward Heath, was able to form a minority government in the sure knowledge that a few months later he would be able to call a second election in which he could hope to secure an overall majority.[2] That path will not be open to whoever forms the next government. Four years ago parliament passed the Fixed-terms Parliament Act, which takes away the right of the crown to call an election at the prime minister's behest. An early election can only be held if two-thirds of the Commons vote for a dissolution, or if the government loses a vote of confidence and no new administration is formed within 14 days. In practice, this means an early election can only be called if the opposition makes it happen or otherwise connives in its instigation. It has been evident for some time that the first-past-the-post system is kinder to Labour than the Conservatives.[3] In 2005, Labour won a majority of the vote despite being just three points ahead of the Conservatives.

By contrast, David Cameron was unable to secure an overall majority in 2010, despite enjoying a seven-point lead. The seats that Labour wins tend to have fewer registered voters and lower turnouts, while the party loses fewer seats to the Liberal Democrats and wastes rather fewer votes piling up large majorities. It is this Labour advantage that explains why the Conservatives were so keen to have the constituency boundaries redrawn on a more equitable basis (at least so far as the number of registered voters is concerned) – and why the Tories' electoral prospects were so badly damaged when the Liberal Democrats scuppered the boundary review in retaliation for the shelving of Lords reform.

Consequently there is at least some chance that Labour could win more seats despite not having the most votes. If the Liberal Democrats remain on their knees – earning no more than their current poll average of 8 per cent of the vote – this is likely to help the Conservatives more than Labour. Nonetheless, the Conservatives will – if other things remain unchanged – need to be something like two and a half points ahead in votes before they are likely to pull ahead of Labour in terms of seats.

Mind you, there is no guarantee that other factors will remain unchanged. Some of the larger increases in the electorate since 2010 have been in Labour-voting constituencies, most notably in the east end of London. Meanwhile, recent electoral history suggests that if there has been only a modest swing to Labour since 2010 (which is what a small Tory lead would imply) some of those Tory MPs who were elected for the first time in 2010 at the expense of an incumbent could have gained enough of a personal vote to hold back a rising Labour tide.[4] But perhaps most importantly, if Labour were to lose, say, 30 of its 41 seats in Scotland as a result of the SNP 'surge' (on which more below) anything more than a one-point Conservative lead would probably be enough to put Cameron ahead. Still, even in this scenario there is some risk that the Conservatives could secure the most votes but not the most seats.

The smaller parties' prospects for power

Which brings us on to the prospects for the smaller parties. On average, February's polls have put Ukip on 12 per cent support, the Greens on 6 per cent. Both have shown some signs of slippage from the especially high levels of support that they were enjoying at the turn of the year, but both are still well ahead of anything like the levels of support they have previously registered shortly before a general election. Yet in neither case does there seem to be much prospect of their performance reaping a reward in terms of seats that is any better than meagre. Despite being particularly popular along much of the east coast of England, even on 12 per cent of the vote Ukip's support looks set to be too evenly spread geographically for the party to be able to pull ahead locally in more than a handful of seats. Meanwhile the hopes of the Greens rest on one or two university seats that have particularly socially liberal and environmentally concerned populations.

By contrast, the SNP's support is concentrated in one part of Britain. Thus while it may represent no more than 4–5 per cent of the vote across the UK as a whole, its current average poll tally of a little over 45 per cent north of the border is enough for it to pull ahead in most Scottish seats. Even less optimistic projections of the nationalists' prospects suggest the party could win 40-odd seats, more than 20 of which would come at Labour's expense, while others suggest it might win as many as 50 seats (with more than 30 captured from Labour).

Such an outcome would not only see the SNP win far more seats than either Ukip or the Greens, despite perhaps securing a smaller share of the vote (an outcome that might also be enjoyed by the Democratic Unionists in Northern Ireland), but it would also see the party overtake the Liberal Democrats in terms of seats. Just how many seats Nick Clegg's party might win on their current 8 per cent of the vote is much-disputed; the party itself is trusting that the local popularity of many of its MPs will enable them to hold their seats despite the apparent strength of the ebbing tide. Even if these hopes are exaggerated, the party has built up enough local strength in some of its seats that it should retain 20 or so even if its nationwide support remains as low as 8 per cent – enough to keep it ahead of Ukip and the Greens but potentially still well short of the SNP.

In short, while first-past-the-post looks as though it will treat some of the smaller parties harshly (as it is supposed to do), in other cases it will fail to do so. Those smaller parties whose vote is geographically concentrated could potentially profit from the system – with the consequence that MPs for parties that only contest the election in Scotland (or Northern Ireland) could find themselves with greater parliamentary leverage than those representing parties that have won votes across a broad swathe of England.

Of course, none of the smaller parties will have any kind of leverage unless there is a 'hung' parliament. Whether that happens will depend in part on how close the result is between Labour and the Conservatives and in part on how many seats are won by other parties. So far as the latter is concerned, the implication of the analysis above is that even with the decline in the Liberal Democrats' fortunes and the potential squeeze on the Commons representation of Ukip and the Greens, that 'other parties' tally is unlikely to be much lower than the 80 or so it has been at each of the last four elections. If that does prove to be the position then, given the potential losses north of the border this implies, Labour could well need a five-point lead to be able to secure an overall majority. Meanwhile, the Conservatives are unlikely to secure an overall majority without at least a seven-point lead – and that presumes that no significant rise in Liberal Democrat support above the 8 per cent mark. Labour has not been five points ahead in the polls since spring 2014, while the Conservatives have not been seven points ahead since June 2010. Both targets are miles away from the narrow one-point lead that Labour has held (on average) over the Conservatives during the last four months. It would appear that neither party is likely to achieve an overall majority unless it fights a particularly effective election campaign.

Negotiating a hung parliament

We thus face the possibility of a hung parliament in which most seats are held by the party that came second in votes and the greatest leverage among the smaller parties is held by the party that came sixth at the ballot box. That would not be an easy outcome for advocates of first-past-thepost to defend, and would seem bound to spark a significant debate about the future of the electoral system. But if we consider the possible implications of anything like such an outcome for the formation and stability of the next government, we can see that that debate may well range more broadly across Britain's constitutional settlement.

First of all, there is no guarantee that the next prime minister will be the leader of the party with the most seats – even if it does also have most votes. As it happened, while being willing to talk to both sides, the Liberal Democrats in 2010 proved able to strike a coalition deal with the largest party, the Conservatives, while being unable to do so with the second biggest, Labour. That agreement with the Conservatives, and the difficult parliamentary arithmetic that would have underpinned a minority coalition deal with Labour, made it relatively easy for Nick Clegg to decide to put David Cameron into Downing Street. But even if the Liberal Democrats were once again able to come to some agreement (either in the form of a coalition or a confidence and supply arrangement) with the largest party, be it Conservative or Labour, they may no longer have the parliamentary strength to take that party past the 326 mark. Instead, that ability may lie uniquely in the hands of the SNP.

The SNP's position, together with that of their Plaid Cymru and Green allies, is not so even-handed as that of the Liberal Democrats. They have ruled out doing a deal with the Conservatives, and that sentiment is reciprocated. In contrast, they are willing to back a Labour government, albeit only from the backbenches. But this support is not being offered without strings attached: demands include abandoning any attempt to replace the Trident nuclear programme and an 'end to austerity'.

Two implications flow from the nationalists' potential strength and negotiating standpoint. First, Labour would appear to have more potential allies than the Conservatives. That could enable the party to form a government in circumstances that the Tories could not, even if Labour has fewer seats of its own. Second, however, actually reaching an agreement with the nationalists will not be straightforward. That in turn could create circumstances in which the Conservatives are able to hang on to the reins of power even without commanding a majority of seats.

We need to remember that strictly speaking there is no obligation on an incumbent government to resign unless and until it has lost a vote of confidence in the House of Commons. When Stanley Baldwin lost his majority in 1923 he forced Labour and the Liberals to vote down his king's speech before resigning and paving the way for Labour (as the second-largest party) to take the reins of office with the Liberals' support. If Cameron were to conclude that there was a chance that Labour might be unable to strike the deal or deals needed to win a confidence vote, he could see if his opponents are willing to take the risk of voting him out of office, for fear that if they were to do so and subsequently fail to put an alternative government in place they would, under the requirements of the Fixed-term Parliaments Act, precipitate a very early second election indeed.

The unsettling effect of minority government

Irrespective of how it might happen, there would doubtless be a hue and cry about the fact that the leader of the party with most seats was not the occupant of Number 10. But once first-past-the-post fails to deliver the largest party an overall majority, it is inevitable that the link between the voters' choice and the partisan colour of the government becomes more tenuous. And in any parliamentary democracy it is ultimately the ability of the government to command the confidence of the legislature that determines both its formation and survival.

That link between voters and the government could also become more tenuous if, as should by now be evidently possible, the election is followed by the formation of a minority government. The nationalists are only willing to contemplate a confidence and supply arrangement, and the same is true of the Democratic Unionists (and Ukip); meanwhile, if the Liberal Democrats were to suffer serious losses then they might start to wonder whether they can afford to suffer the slings and arrows of government unpopularity for another five years (assuming they even have the numbers to form a majority coalition with either side). In any case we should not presume upon the willingness of either Conservative or Labour backbenchers to suffer the 'constraints' of coalition. 'Going it alone' might seem more attractive to them too.

Regardless, forming a minority government – as happened in 1974 – would not be without its risks and difficulties. It will not be certain that it can pass its legislation, especially given that MPs of all parties are more inclined these days to rebel against their own party whip. Yet no matter how difficult the position in which it comes to find itself, that government will be unable to call an election of its own volition. Instead, the Commons might opt to vote it out of office and put an alternative administration in its stead without reference to the voters at all. Indeed, experience with fixedterm parliaments in other countries suggests that this is what sometimes happens when a fixed elections rule is in place.[5] And after all, even under the old rules, in 1905 and 1931 new administrations took the reins of power without an election, albeit that on both occasions the electorate was consulted shortly thereafter.

Such a development would certainly offend some people's conceptions of how Britain's democracy should work.[6] Yet the previous system can be criticised too. After all, incumbents are more likely to be re-elected in countries where prime ministers can go to the polls at a time of their own choosing, suggesting that such discretion can be used to achieve electoral advantage.[7] And if elections can no longer be relied upon to produce a clear outcome, then they may not provide an unambiguous answer to the 'who should govern' question that the electorate might be asked to address after a government has fallen.

Finally, we cannot forget the debate about devolution and 'English votes on English laws', for it is the governance of England that will primarily be affected by the partisan composition of the next government. That in itself may raise questions about the merits of having a Scottish (or a Northern Irish) party potentially holding the balance of power. It is, in truth, a possibility that is always more likely to arise with an electoral system that is kinder to parties whose support is concentrated in a particular geographical corner of the UK. But even if the nationalist parties do not prove to be kingmakers, there is no guarantee that the next government will have won the highest number of English seats. A narrow Labour lead in terms of seats could well be based on the party's relative strength in (substantially over-represented) Wales along with whatever spoils it manages to retain in Scotland. It would be the first government to find itself in that position since the Labour one in – you guessed it – February 1974. But what might have been tolerated then, before the advent of devolution, may be less acceptable now.

A two-party system of alternating majority government has long been one of the key foundations upon which the UK's constitutional edifice has rested. It has provided a clear link between the voters' choices in the ballot box and the partisan colour of the government. But if the electoral system fails to sustain that link we are likely to find ourselves facing a serious debate about how the country's electoral and constitutional process can and should operate. What perhaps is less certain is whether that exercise is likely to provide any clearer answers about what should be done than it did four decades ago.

John Curtice is professor of politics at Strathclyde University and a columnist for Juncture.

This article appears in edition 21.4 of Juncture, IPPR's quarterly journal of politics and ideas, published by Wiley.


1 See Finer S (ed) (1975) Adversary Politics and Electoral Reform, Anthony Wigram. ^back

2 In practice, the ruse proved less successful than Wilson had hoped. In October 1974 Labour won an overall majority of just three seats, which proved insufficient to withstand the vicissitudes of death, disappearance and by-election defeat, and eventually forced the government into a 'pact' with the Liberals. ^back

3 Curtice J (2010) 'So what went wrong with the electoral system? The 2010 election result and the debate about electoral reform', Parliamentary Affairs, 63: 623–638. ^back

4 Curtice J, Fisher S and Ford R (2010) 'Appendix Two: An analysis of the results', in Kavanagh D and Cowley P, The British General Election of 2010, Palgrave Macmillan. ^back

5 See Schleiter P and Isar M (2014), 'Fixed terms parliaments and the challenges for governments and the civil service: a comparative perspective', Political Quarterly, 85: 178–186. ^back

6 See Bogdanor V (2011) The Coalition and The Constitution, Hart. ^back

7 Schleiter P and Tavits M (2014) 'The electoral benefits of opportunistic election timing', conference paper, Annual Meeting of the Midwest Political Science Association, Chicago, 3–6 April 2014. uk/Lists/Events/Attachments/397/2014Apr-Abstract-Timing_Schleiter_Tavits_MPSA_2014.pdf^back