The YouGov/Sunday Times poll had some questions trying to tease out people’s perceptions of who has the best claim to be PM in a hung Parliament. This is, obviously, not necessarily the same thing as who will be. Much of the discussion I’ve seen on this has been at cross purposes – some people rightly saying that the leader who can command a majority in the Commons has the constitutional right to be PM, others saying that in circumstances X, Y or Z or with party A, B or C that may be seen as illegitimate. These two things are not contradictory – it is perfectly possible to have a situation where a leader has the perfect constitutional right to be Prime Minister, yet is seen as illegitimate by the public. If the study of public opinion tells you anything, it should be that public opinion is quite often wrong. A good example is Gordon Brown in 2010 – remaining as PM while negotiations took place was quite clearly his constitutional duty… but it didn’t stop him getting flak for “squatting” in Downing Street. Public opinion on the legitimacy of who becomes PM won’t make any difference to who gets the invite from the Palace, the maths will decide that, but it may make a difference to how that government is perceived by the public in the longer term.

On this front, by 47% to 26% of people think that the biggest party has the best claim to form a government, even if other parties collectively have more seats. If there is a difference between the party with the most seats and the most votes, by 43% to 29% people think it is votes that should matter.

Asked about whether parties should try to go it alone or form a coalition there is an interesting difference. Should the Conservatives find themselves the largest party then 58% of Tory voters think they should try to strike a deal with other parties to get a majority, 29% think they should try to go it alone. Should Labour find themselves the largest party the figures are much closer – 44% of their voters think they should try to strike a deal, 39% think they should try to go it alone. YouGov then asked what the other side should do in those circumstances… in both cases, the balance of public opinion is that oppositions should give a minority government a chance. If the Conservatives try to go it alone, 32% think the other parties should vote to bring them down, 40% think they should be given a chance. The figures are almost identical for a minority Labour government, 30% think the Tories should just vote them out, 39% that they should give them a chance.

The polling on all these questions will likely be transformed completely next week when the numbers are known and these questions become opinions on a Cameron government, a Miliband government or whatever, rather than hypothetical situations – these aren’t set in stone. I expect many respondents who say largest party should form the government might change their answer in the event largest party was X or Y. The point us how the parties behave next week, whether they are seen as being in the right and behaving in a responsible way will have an impact on the public’s perception of them.

Both the YouGov/Sunday Times poll and the Survation poll asked people who watched the Question Time leaders special earlier in the week who they thought had won – both found Cameron clearly ahead. YouGov had Cameron winning by 42% to Miliband’s 26% and Clegg’s 13%, Survation had Cameron winning on 38% to Miliband’s 24% and Clegg’s 9%.

As well as the YouGov/Sunday Times poll there was also a separate YouGov poll for the Sun on Sunday. This has topline figures of CON 34%, LAB 35%, LDEM 9%, UKIP 12%, GRN 5% so is also bang in line with YouGov’s pattern the parties being roughly neck-and-neck. The poll included a question on people’s preferred coalition/deal which showed a very even split, the same as we’ve seen in many other polls – Con/LD 21%, Con/UKIP 18%, Lab/LD 20%, Lab/SNP 16%. However they also asked which coalition people think would be worst, which produced a much clearer result – Lab/SNP 39%, Con/UKIP 32%, Con/LD 6%, Lab/LD 4% – people fear the SNP and UKIP’s influence on government, the poor old Lib Dems are seen as quite benign.

1,586 Responses to “More from the Sunday polls”

1 2 3 32
  1. So I was right Oldnat about there being a separate Sun on Sunday poll!

  2. Thanks Anthony. Another new thread. You are having to work overtime with so many polls flying around. I’m a bit baffled as to why there was a separate Sun on Sunday poll in addition to the daily YouGov.

    It does seem like its back to neck and neck and averaging out at 33/33 each. Every time either Lab or Con have moved ahead they have been pulled back the following week.

    I love politics and this election is fascinating but I won’t miss the deluge of polls.

  3. BigD

    You were!

  4. James Peel

    “The rise of the SNP has been a godsend to the tories, and Cameron in particular.”

    I would agree with that.

    However I can not believe Cameron, will act in the same way as he did on the morning after the no vote in the Scottish referendum.

    If he has no majority , but he calls any other possible government illegitimate, he will be bringing parliament into a more dangerous territory than it has been in many peoples lifetimes.


    John Curtice on polls and a wee bit of Indyref2 talk at the end:

    “However, it appears that as a result of this campaign voters are increasingly coming to the view that a second referendum will be held sooner rather than later – even though at the same time support for holding an early referendum has fallen. Back in March 45% of Scots said that there should be a referendum within ten years; now that figure stands at just 36%. But whereas in March only 40% thought there actually would be referendum within that time scale, now as many 54% reckon there will be. The decision that Nicola Sturgeon will need to make next year about whether or not to include a promise of a second referendum in the SNP’s Scottish Parliament election manifesto is evidently not going to be an easy one.”

  6. “On this front, by 47% to 26% of people think that the biggest party has the best claim to form a government, even if other parties collectively have more seats”

    Unsurprisingly, Scots responses worked the other way. 50% thought the group with most seats should form the government, while only 33% thought being the largest party bestowed that right.

  7. Good early Afternoon All, from Bournemouth with blue skies now as we wait for tomorrow’s pier to pier show of the Championship/Division One Title by Eddie Howe’s team, moving from Bournemouth East to Bournemouth West.

    Big swing in the next few days, I think; plus or minus three per cent, in England, and in Scotland too. I think.

  8. Back in March 45% of Scots said that there should be a referendum within ten years; now that figure stands at just 36%. But whereas in March only 40% thought there actually would be referendum within that time scale, now as many 54% reckon there will be.

    Yes. This is the most interesting nugget of information in all of the virtual forests of information abounding this morning.

  9. @oldnat

    Rather amusing Scots think that way, as with all those Labour seats likely to go SNP, the Tories look likely to be the largest party.

    Perhaps they will twig this contradiction in the remaining few days, and realise that a vote for SNP returns a Tory government.

  10. @dez,I think the rise of UKIP could be a godsend to EM in the target Con Lab marginals,given UKIP have far more previous Tory voters .This is where the election will be won ,If bluekippers go home DC wins,if not he is done for

  11. “Perhaps they will twig this contradiction in the remaining few days, and realise that a vote for SNP returns a Tory government.”

    I think this interesting. I have always thought the idea that the SNP would win nearly every seat a bit fanciful. I expect many people who say they will vote SNP will do something different. I still expect labour to hold about 6-10 seats in Scotland.

  12. Sadly, I think this question of what coaltions and/or alliances might or might happen after the election has completely derailed this GE. It’s no longer about policies and what sort of economy, NHS, immigration etc.etc. policies people want but all about what happens after the vote.

    I listened to John Pienaar’s programme on Five Live this morning and it was just obsessed with this and whenever anyone says that Ed Miliband has ruled out any formal alliance with the SNP, it is just scoffed at by Tories and “independent” commentators. Even Danny Alexander was forced by the newspaper “experts” to have to support what Ed Miliband said after he was asked would the LibDems join a coalition with Labour and SNP. He said there isn’t going to be such a coalition as “we have to take Ed Miliband at his word.” But that wasn’t good enough as they still wanted him to say what the LibDems would do in that event.

    Then they wanted to know what the LibDems would do if the Tories enetered a coalition with UKIP or with DUP or with UKIP and DUP. It was never ending. There was no discussion at all about LibDem policies. Earlier on there was a similar discussion of “possibilities” with Chuka Umuna. No discussion of policies.

    Are we all now expected to vote on the basis of who might ally themselves with who after the elction rather than for which party we think has the right policies? I find it very dispiriting that this is what it has all come down to now.

  13. Chris Lane

    “Big swing in the next few days, I think”

    Today’s YG –

    No chance of changing mind – GB 79% : Sco 85%
    Might change mind – GB 7% : Sco 9%
    Don’t know – GB 11% : Sco 3%

    Lowest level of decision made in London (73%) : Mid/Wales (75%)

  14. Might or might not. Umunna. Where’s the edit button!.

  15. Just for clarification – the main Northern Ireland party in the Commons is the DUP not UUP. The DUP overtook the UUP several years ago. The UUP currently has no MPs – but may win 1 on Thursday.

    Despite DUP frustration at Tory tactics over SNP, they still seem more likely to go with Con/LD deal – as the Tories are prepared to meet their demands with the possible exception of Bedroom Tax. However as LDs are no longer wedded to the Bedroom Tax, a review of the policy would probably be enough to get DUP on board. Based on Kellner’s forecast this morning that will get DC over the line without even needing UKIP.

  16. Statto

    I suspect you read the numbers the wrong way round. More Scots were saying “being the largest party” should NOT bestow legitimacy.

    In other words Murphy’s argument is only believed by those intending to vote Labour (plus the Tories).

    That shouldn’t amuse Lab folk.

  17. @ Norbold

    I think this is because the general positioning was finished early last week. I doubt that it is even about small clusters of voters anymore. I think it is now mainly about tactical voting.

  18. Not that I’m complaining however surprised how much in Cameron’s favour the QT questions came out; also surprised how poorly Clegg faired in that question.

    The ‘which coalation would be the worst’ is a fairly clear picture and jusifies the Tories continuing with the Lab/SNP line, even if its a looser arrangement rather than coalition. Unless the polls are completely wrong, Lab will need SNP support

  19. …there was also a separate YouGov poll for the Sun on Sunday. This has topline figures of CON 34%, LAB 35%, LDEM 9%, UKIP 12%, GRN 5%

    Blimey. I missed this. Was this result even tweeted last night? Can’t find anything by @SunPolitics.

    [I’m not sure if the @SunPolitics account is just the weekday edition team – for the Sun on Sunday stuff it’s best to follow @DavidWooding, the Sun on Sunday’s political editor – AW]

  20. I always read the comments late, often in one go, so I’m well behind on this. But, seeing as the new post atl is on what happens after the election, I was struck by two predictions made last night, which seem perfectly sensible and highly plausible – but if they happened would lead to two very different scenarios, so are worth pausing over

    Lab 277
    Con 271
    SNP 47
    Lib Dem 29
    DUP 9
    Sinn Fein 5
    UKIP 3
    Plaid 3
    SDLP 3
    Green 1
    Lady H 1
    Speaker 1

    Lab 274
    Con 276
    SNP 51
    Lib Dem 23
    DUP 9
    Sinn Fein 5
    UKIP 2
    Plaid 4
    SDLP 3
    Green 1
    Lady H 1
    Speaker 1

    At first glance, they both seem to fall into ‘Camerson’ can’t be PM category, which is true. But if @RIVERS10 is proved correct, I would expect a relatively early resignation. Firstly, Lib Dems may insist on talking to Labour first, but of perhaps more important would be the fact that (I think): a) both Labour and SNP would signal an early determination to vote down any Tory-led Queen’s Speech, and their combined 324 means that no combination of others would defeat them (Sinn Fein, as always, key in this respect), b) Miliband would signal early his intention to govern alone, with a large minority, if no other party comes in, and basically challenging SNP to dare vote a Labour Government down. A Lab-Lib collation may then be formed, but with Labour in a massive position of strength. It may work, or it may all fall apart, but either way Miliband would be PM of the largest party. The next GE might come quickly, but if Labour lost a vote of NC, I’d suspect they’d be a quick election as it would probably be clear an alternative would not work. Remember, Miliband would not have to resign as PM as the 14 day window to form a new Gov is open. So, this scenario would probably mean no Tory QS, but a Labour one; either a shaky gov lasting, or a quick GE with Miliband as PM of the biggest party – not a factor to be taken lightly in nay campaign.

    @COUPER2802’s prediction would lead to a very different scenario. Yes, the anti-Tory block is even bigger, with Lab+SNP having 325 seats, but Cam would be within his rights to go for a minority administration, or continued Con-Lib, safe in the knowledge that if he loses on the QS/NC he and the Tory press could scream ‘squatting’ in Downing Street if Miliband takes over on the basis of being the second largest party, and with SNP propping him up. They’d undoubtedly do the same in the first scenario, but it would be much stronger if it could be argued that Lab had denied the ‘winners’ a chance to govern alongside the SNP. It wouldn’t surprise me in this scenario if there is a Con QS, but early defeat on QS/NC, but that Miliband accepts that he cannot govern – therefore another election quickly with Cam as PM. If Miliband does take over, queue howls of protests. This is where Fixed Term is a real issue, as in this scenario, the ideal thing for Miliband would have been Cam to resign after losing QS, Miliband take over and go for an instant election. He can’t in this scenario, meaning he’d every have to run a minority collation ‘propped up’ by SNP, which I honestly think would spell disaster for Labour in the future, or see another early election with Cameron as incumbent.

    I’m a Labour supporter, but I genuinely think the ‘legitimacy’ issue is a major one if Lab is second biggest party, and they with Libs do not have the vote. Ruling out any ‘deal’ with SNP leads him open to hypocrisy or squatting charges, unless Labour has the biggest number of seats.

    Sorry for the length of this post/essay!

  21. James Peel

    “I expect many people who say they will vote SNP will do something different.”

    Indeed. As may people who say they will vote Lab, Con or LD

    “I still expect labour to hold about 6-10 seats in Scotland.”

    That may well happen. There are a number of seats where I anticipate the margin of victory being very small.

    A small variation in turnout could tip these seats one way or another.

  22. “Public opinion on the legitimacy of who becomes PM won’t make any difference to who gets the invite from the Palace, the maths will decide that, but it may make a difference to how that government is perceived by the public in the longer term.”

    Thank you, AW!

    Hopefully that will make this thread void of arguments to the contrary.

  23. The huge problem with the late late swing back theory is that we have no way to differentiate between that and general polling error. If all the polls are underestimating party X we would see exactly the same results.

    (The late late, or ‘strong’ swingback is the one which only turns up in he ballot box, after the final opinion polls have been put to bed. It is the b*stard love child of everyone who insists that in the privacy of the polling station people won’t be able to bring themselves to vote for a party that would do A, B, or C)

  24. Interesting to see the seat projections tending to swing towards Labour today. On the numbers that I’ve seen, it looks clear that Cons could not form a government. This may explain the newspaper reports that Cons are telling journalists that Ed should ‘do the decent thing’ if he comes second in seats. It doesn’t sound like a party that is expecting to win sufficient seats to govern.

    Against this, have we had ‘the Sheffield moment’? Tories are trying to say we have, with Ed’s pledge stone. Storm in a tea cup, or gamechanger? I groan at these nonsensical gimmicks, and wish that politics wasn’t run by teenagers.

  25. OldNat

    I think at this stage of the election, it’s a bit of a nonsense to ask these questions about what confers legitimacy, since most people understand the scenarios at least well enough to know what favours their party – I’d expect more Labour and SNP voters to say being the largest party really isn’t that big a deal, whereas more Tory voters would say that being the largest party in terms of seats is absolutely sacrosanct and any other outcome would see the oceans boiling and plagues across the land.

    We’re going to hear some A-grade nonsense on these subjects from about 4am on Friday next week. I’m expecting an early Cameron declaration of victory in the style of his silly EVEL speech the day after the referendum, just to kick things off.

  26. @OldNat

    I’m not seeing many big swings, just some wobbles <0.5%, basically nothing much happening. If there is any swingback to be predicted from this, it's going to be pretty tiny.

    That's 150,000 votes or so. So a more modest swing is not a lot of votes on that scale. Wouldn't if be fun if one main party were ahead on the popular vote by single figures? That does not seem so absurdly impossible now.

  27. TingedFringe

    I’m not sure whether there were any comments about who would or wouldnt go to see the queen however the ‘legitimacy’ argument, certainly in my mind, was always about public opinion/perception.

    If any PM isnt viewed as being legitimate then they really wont last long, unless all their policies etc come good and the economy motors. One whiff of trouble and they’d be toast….which makes it one hell of a risky game for whichever party they’d represent. Its the sort of thing which could taint a party for years, similar to how some of Thatchers policies still taint the Tories in part of the UK.


  28. @Magpie

    If it’s anywhere near close I suspect we won’t see any declarations of victory. Especially from the back of a train in the style of Dewey…


    The numbers may also play to the tone of the negotiations, first mover advantage can be crucial on occasion.

  29. The fact that the Sun on Sunday YouGov had a 1 % Lab lead, and the Sunday Times had a 1% Conservative lead — when both polls were done at the same time — says it all about how much significance we should invest in an individual 1% lead on a poll.

    Bottom line: two main parties neck and neck. No swingback.

  30. @AdamB

    Being viewed as legitimate is an issue that has happened many times in recent UK history: Brown, Major, Douglas-Hume all becoming PM without winning the election. In the end it doesn’t cause the Government to fall.

  31. @MatthewG

    Note that with the Rivers10 projection Con+LD = 300 and with the Couper2802 one Con+LD = 299. So it’s basically LD and Con swapping seats around or not. What a difference in perception one seat makes.

    However – a government does not have to be “legitimate” – it just has to be able to survive (or be able to avoid) a no-confidence vote. To me 299 or 300 isn’t enough, even with the DUP helping out.

  32. ProfHoward – actually the Sun on Sunday one was done first, but the point about how little meaning a one point lead has is correct anyway.

  33. @oldnat

    I didn’t explain that well. The contradiction in my mind is that the “largest group” will be one and the same as the largest party. There are no coherant groups other than these.

    Labour and the SNP are bitter enemies. partly because a large part of the SNP are ex- Labour who left in frustration, and those that remain in Labour see nationalism as being entirely incompatible with socialism.

    Interesting article in the Scotsman online
    referring to why former first minister McConnell allowed the SNP to form a minority govt at Holyrood. In short, it was because they were the largest party, had been perceived to win and any other government would have lacked legitimacy.

    Much the same arguments being made now.

  34. TheSheep
    As you point out, becoming PM without winning an election is quite normal – as we’re not a presidential system, its party based. I havent checked the numbers but i’m guessing that in most cases the PM was still leader of the largest party.

    Someone becoming PM when they are a clear second place in an election and where there is a materially larger party in Westminster is quite different.

  35. “Public opinion on the legitimacy of who becomes PM won’t make any difference to who gets the invite from the Palace, the maths will decide that, but it may make a difference to how that government is perceived by the public in the longer term.”

    But the problem with this is that an ‘invite from the palace’ isn’t a straightfoward thing. The palace will want to avoid any controversy. And remember, if Tories are biggest party, Cameron is well within his rights to argue he is the only legitimate PM, to call on other parties to support/abstain in a QS on the grounds of ‘national interest’, and if he loses howl about a vile plot between a Lab and SNP leading to an ‘illegitimate’.

    If Tories are biggest party, then the ‘call from the palace’ will only come if and when Cameron decides he can’t possibly run the country, and the politics of ‘legitimacy’ will come into this. The ‘palace’ isn’t an umpire, but acts reactively to events.

  36. @OLDNAT That may well happen. There are a number of seats where I anticipate the margin of victory being very small.

    The interesting thing here is the scale of the swing. I mentioned on a previous thread that it’s likely the ‘biggest ever swing in a GE seat’ will not just be broken, but broken by a number of SNP MPs.

    The current predictions only show a handful of marginal seats compared with the tight races in England, however with such a huge change in the vote, it does seem logical that the change is more volatile.

    In order to start retaining more seats Labour need to reduce the swing by 5%+, the kind of figure that would be ridiculous in England – BUT when we’re talking about predicted swings of 25% already, it’s not quite so fanciful.

  37. I’m re-posting this from the previous thread:

    I’ve been doing a bit more reading about precedents set in previous elections which were as close as this one is expected to be.

    The scant precedent which exists is: A government which has failed to pass its own QS (Queen’s speech) abstains, rather than voting against the opposition leader’s QS. This precedent could explain why Ed M can say that he won’t need the SNP to get him into government. Remaining there is, of course, an altogether different ball game.

    Gary O (& me) on the previous thread are wondering whether the cabinet office has advised the Conservatives of this QS precedent & whether the 1922 committee will advise the Tory backbenchers to follow it.

  38. @KeithP

    My point is exactly to try to think through the different implications of tiny changes. Yes the Con/Lib bloc might not be different, but if Cons lead Lab, then the behaviour of the Conservatives I think will be massively different, as they will be jockeying for political position – and legitimacy comes into this.

    Legitimacy is a political, not a ‘constitutional’ issue.

    I myself see no problem with a Lb+Lib+SNP coalition, but Miliband has, in my opinion, foolishly ruled it out.

  39. Magpie

    Agreed, but that’s the question that the Sunday Times decided to pay YG to ask, and Anthony dutifully reported above.

    Some may think that this is part of a Tory strategy of delegitimising the constitution of the UK (such as it is) for party advantage.

    The argument that the largest party should (God knows how if the parliamentary arithmetic doesn’t work) form the Government seems designed to remove parties outwith the big two (and their voters) from participation in the democratic process.

    I would suggest that the difference between the GB and scottish figures, that I quoted above, are the result of the extensive discussion of the issue in Scotland for the last few months.

  40. Oh God, what have I started?

  41. @Amber Star,
    Very interesting. I think this is key, and wonder whether we might see a Con QS, defeat, then Lab in power (if Cons biggest party). If Lab win most seats, I expect a pretty quick concession, although Cameron will probably remain as PM for a bit.

    Staying as PM until a new gov is formed is not so much Cameron’s right as his duty… as it was with Brown in 2010.

  42. I re-read Adonis’ Five Days in May the other day, pretty much text-book how not to form a coalition…

  43. Matthew G

    Clegg has also ruled out working with the SNP so unless he loses his seat then Lab+LD+SNP is a non-starter.

    re the talk of Lab/SNP marginals, to a certain extent this is noise given Lab will likely work with the SNP via whatever arrangement. The reason its not completely academic is obviously due to the Con v Lab gap.

    However the main decided in teh election will surely be Con/Lab marginals given they effectively count double (one party goes up one, the other goes down one), rather than in Con/LD or Lab/SNP marginals


  44. Labour has been very clearly testing the softness of the Green and LibDem vote for tactical voting purposes in the last two days.

  45. @ADAMB Someone becoming PM when they are a clear second place in an election and where there is a materially larger party in Westminster is quite different.

    It’s how coalitions work all over the world. It’s actually quite depressing that the demonising of the SNP has seemingly persuaded some of the public that not every MP’s vote counts.

  46. Laszlo
    Have they? I’ve been surprised that EM has been in Scotland – I would expect Cameron and Miliband and all other bigwigs from the two parties to spend their time in England marginals between now and Thurs

  47. Prof H/Anthony

    I still don’t understand why YouGov did two polls last night. I appreciate the Sun on Sunday polling was conducted earlier but why two polls?

  48. Statto

    I’ve heard McConnell on that argument before.

    If McConnell had been able to get the support of the LDs, then they would have been able to form a larger minority government than the SNP.

    However, the LDs weren’t prepared to a deal with either of the big two parties.

    At the time, the consensus was that a minority government couldn’t survive anyway, so sitting back and watching the SNP collapse was considered a wiser strategy. Then, at the extraordinary general election ….. etc etc.

    When things turned out differently, it might be thought that McConnell’s best plan was to explain the stance he had taken as being honourable.

  49. Matt M
    SNP aren’t being demonised. Its quite irrefutable though that they only represent one geographic part of the UK, that their raison d’etre is breaking up of the UK and their policies require materially increased spending compared to other parties

  50. The sheep

    Agreed, the only way to properly judge a stochastic model is over many elections. The whole field of predictive models doesn’t hang on this one election.

    So far there is a small amount of evidence for a swingback trend (this may need to be tweaked in the cases of enormous swing but still fitting history). This trend might be real it might not and be shown to be a certain amount of noise that in 20 elections time was shown to be a very minor effect.

    You really only need one hypothesis to generate a swingback model.
    Basically it says people sometimes switch their allegiance, but they are more likely to switch their allegiance to a party they have previously voted for than any other party. It’s not very sophisticated but it’s not unreasonable that it’s a better hypothesis that all VI move as a pure random walk from the current VI.

    More sophisticated models would probably try to capture “core votes” etc but it’s a bit of a case of walking before you can run.

    In a way they did get the fundamental movements right, Lab fell from 38 and Con grew from 30. LD failed to recover by much, which is largely to the “no hearing” effect. UKIP did fall but not by as much as expected.

    Comparing the forecast to the nowcasts when labour were sitting on 38% and they will look pretty good. They certainly predicted a hung parliament very early on. To answer do they add any predictive capability, the answer is “some”. Which isn’t bad for a toy model.

1 2 3 32