Stephen Bush over at the New Statesman has written an interesting article about the mountain that faces Labour at the next election. I’ve now had chance to sit down and play with the election results and the picture is as bleak for Labour as Stephen paints – for various reasons, the electoral system has now tilted against Labour in the same way it was tilted against the Conservatives at the last few elections.

Looking at how the vote was distributed at the general election the Conservatives should, on a uniform swing, be able to secure a majority on a lead of about 6%. Labour would need a lead of almost thirteen points. On an equal amount of votes – 34.5% a piece – the Conservatives would have almost fifty seats more than Labour, Labour would need to have a lead of about four points over the Conservatives just to get the most seats in a hung Parliament. The way the cards have fallen, the system is now even more skewed against Labour than it was against the Conservatives.

How did this happen? It’s probably a mixture of three factors. One is the decline of the Liberal Democrats and tactical voting – one of the reasons the electoral system had worked against the Tories in recent decades was that Labour and Lib Dem voters had been prepared to vote tactically against the Tories, and the Lib Dems have held lots of seats in areas that would otherwise be Tory. Those factors have vanished. At the same time the new dominance of the SNP in an area that was a Labour heartland has tilted the system against Labour. Labour had a lead over the Conservatives of 9% in Scotland, but Labour and Conservative got the same number of Scottish seats because the SNP took them all.

Finally there is how the swing was distributed at this election. Overall there was virtually no swing at all between Labour and Conservative across Great Britain, but underneath this there were variances. In the Conservative held target seats that Labour needed to gain there was a swing towards the Conservatives (presumably because most of these seats were being contested by first time Conservative incumbents). In the seats that Labour already held there was a swing towards Labour – in short, Labour won votes in places where they were of no use to them, piling up useless votes in seats they already held.


And, of course, these are on current boundaries. Any boundary review is likely to follow the usual pattern of reducing the number in seats in northern cities where there is a relative decline in population and increasing the number of seats in the south where the population is growing… further shifting things in the Conservatives favour.

736 Responses to “The mountain facing Labour”

1 2 3 15
  1. Your swing analysis suggests that calculating GB VI shares is a complete waste of time!

    If pollsters can find a way to sample the English marginals accurately, then that might help to explain the 2020 result. Knowing that Con have increased/not lost in safe Con seats and Lab have increased/not lost in safe Lab seats is rather a waste of the clients money.

  2. Is Liz on question time too normal to win?

  3. Essentially looks almost impossibly difficult for LAB in 2020 unless Scotland flips back round. Hard to see the Tories losing many of those seats they’ve gained from the Lib Dems. In which case to get a majority LAB would need seats like Kensington, something that would require 1997 levels of support. Not going to happen.

  4. @OldNat

    May not work in quite the same way next time. Going to be a lot less first time incumbents in marginals (apart from in Scotland, of course).

  5. @AW

    In many of the marginals the swing wasn’t towards the Cons as such but from Lab and to a lesser degree Con to Ukip. What saves the Tories was a swing towards then from the LDs. There is virtually no chance of this happening next time.

  6. @AW
    “On an equal amount of votes – 34.5% a piece – the Conservatives would have almost fifty seats more than Labour.”

    Oh well, I was 10 seats out on the previous thread, but without a full dataset available.

    The next question is what would the mountain be if the Boundary Commission recommendations are implemented?

    And what if the full roll out of IER in 2016 causes swathes more people to disappear from the electoral register, as it will?

  7. Jack Sheldon

    Add in first time incumbents as well if you wish.

    The essential point remains the same. Measuring GB VI is a useless exercise, unless a PR system is introduced.

    In a FPTP system aggregating VI from safe seats with ones where change might occur, conceal, not enhances, our understanding of change.

    I’m not sure whether the pollsters sold the idea to the press, or vice-versa, but as a research methodology it has no value other than to get folk on here excited that there has been a change in a meaningless number.

  8. Of course if there’s one thing this election has shown is that results are often a lot more cataclysmic than we all expect! If the LibDems recover strongly (which as a LibDem I very much hope!) and take maybe 30-40 seats off the Tories, if the SNP turn out to be one-term wonders, if Labour break through in the marginals… all big ifs, perhaps, but who would have predicted any of the events of this election five years ago.

  9. I suggest that YouGov dispenses with geographical crossbreaks and instead brings in new crossbreaks based on categories of seats based on their degree of marginality and type.

    Scottish seats could be classed under “Other”.

  10. In 13 years in power Labour had countless chances for electoral reform:

    1. a Lib-Lab pact to make the Libs a party of government and then lock the Tories out forever
    2. follow the recommendations of the Jenkins report with that clear majority and mandate twice
    3. even use the honeymoon in 2001 for a referendum on AV+.

    Shortsightedness and greed got the better of the “reformers” in New Labour. You reap what you sow. Enjoy the wilderness forever.

  11. I just can’t see how labour can get a majority being so weak in the populous south.

  12. A Labour-SNP informal pact seems the most likely government, but there isn’t remotely enough evidence to make a prediction at this stage.

  13. Be interesting to see how the leadership contenders ,and their mentors,face up to this challenge.

    2025 (or will it be 2023) looks a long way off.

    EU referendum could also give Ukip a significant boost .

    Labour campaign for electoral reform likely to see renewed interest.

  14. Bill Patrick

    That might work in a Federal Senate.

  15. I doubt Electoral Reform will be on the agenda in this parliament. The Tories will probably oppose it and would likely find themselves joined in the lobby by the SNP (for the one and only time in the whole parliament), who won 56/59 seats on approx. 50% of the vote in Scotland. The Nationalist turkeys won’t vote for Christmas.

  16. Paul D

    Even on the most basic of glances, a Private Member’s Bill introducing PR for Westminster would stand less than a snowball’s chance in hell of passing.

    Even if the SNP were as cynical as you suggest, they aren’t stupid.

  17. @Paul D

    From 4th May

    The SNP would vote to introduce proportional representation for Westminster elections, Nicola Surgeon has confirmed.

    The Scottish First Minister said that she supported electoral reform despite her party being set to do very well out of the current system.

    “I believe strongly in proportional representation. I believe there should be a direct relationship between the percentage of votes a party wins and the percentage of seats they win in whatever parliament the election is for,” she said. “The polls suggest that my party might do well under first past the post on Thursday but I think if you believe in something in principle you should believe in that regardless of whether your party benefits from the current system.

    “I support PR in principle, it’s in our manifesto, and the SNP would vote for it.”

  18. @ Bill Patrick

    Any kind of involvement with the SNP will be met with the same fear campaign as it was this time. The people of England seem to be most unhappy at the thought of the SNP tail wagging the Labour dog, even if it is an exaggeration.

    Labour need to be very careful in the selection of their next leader, because a London metropolitan will not help them in the North of England nor Scotland, it probably won’t help in the South either. Something tells me that in their political correctness they won’t be able to stop themselves voting Chukka Ummuna in, especially as they’ve built him up to be ‘the British Obama’. Possibly as bad a choice as Michael Foot was.

    Meanwhile in Scotland Nicola Sturgeon is promising impossibilities to the Scots and then blaming the Tories when they can’t be delivered !

    Labour really are stuck between a rock and a hard place

  19. “Labour really are stuck between a rock and a hard place”

    If you select Ukip on their interactive map thingy you can see they have a high spot centred around Yorkshire.

    Perhaps Labour’s national problem is somehow related to that.

  20. @ CMJ, OldNat

    It would be very, very hard, but doable.

  21. @Robin Hood (fit)


    Congratulations! I posted an unprompted comment a few days ago acknowledging your achievement.

    I was always sceptical about seeing swingback in this cycle and before the exit poll I firmly expected to see you eating humble pie. But yes … the 2010-2015 cycle can now be added to the tallies of those fitting the pattern (though the pace of the recover seems to vary remarkably from cycle to cycle).

    From your position of having made a successful prediction, I wonder if you would now be willing to spell out exactly what ‘swingback theory’ states.

    At the end of your comment in the previous thread you acknowledged that there hadn’t been any sign of swingback to the closest equivalent of the ruling party in Scotland. Nor was there swingback to the junior coalition partners overall.

    So what is the precise theory that has been ‘proved correct’ by the election outcome?

  22. It doesn’t matter how big the mountain is for Labour.
    When they pick Chukka Umunna the ethnic block accompanied with white middle class guilt will be enough to see them over the finish line.

    And everyone will be like OMG first black PM, omg history, omg Obamania, omg bigger swing than Blair, omg Chukka so cool, omg Chukka so sexy…, David Cameron stale and pale

    Such is what politics is reduced to nowadays.

  23. Spreadsheet featuringLabour targets for the 2020 election under the current boundaries.

    Note that the swing required under UNS for Labour to win a majority is now 8.7% instead of 4.7%. If an assumption is made that Labour won’t win seats back from the SNP in Scotland, that figure rises to 9.4%, which is only 0.9 points lower than the national swing recorded by Labour under Tony Blair in 1997:

  24. @ Anthony Wells

    There are many dubious points in this post. Labour’s difficulties is not one of them, but projecting current voting patterns (even if we discount the minor issues of the referendum and Scotland) into the future (you know Greens, LibDems, Labour, UKIP). I thought it was a kind of an issue last time.

  25. Are we back to UNS? It seems like.

    Any analysis of the churn in this election?

    Even if it’s meaningless with the boundary changes?

  26. Laszlo

    “Are we back to UNS? It seems like.”

    Not for GB. That was wrong in 2010, and even more wrong now.

    It may apply in E&W.

  27. @ OldNat

    That was my point. It won’t apply (unless miracle is brought in by the government) in E&W either. The Conservative vote will likely be chipped from all directions. The magic that worked against EM is unlikely to be repeated.

    It is, of course, still possible that the left of centre vote is split.

  28. Sorry, the question about the UNS point meant to be sarcastic.

    We just had an election when a party was effectively wiped out because of the possible perception of another party. Two parties had massive voting shares with 2 seats, we also had strangely split constituencies with little visible pattern. We also have the replacement of 2 (or three) party leaders, and before the next election we will see the replacement of a third.

    And a few more minor issues (not last the new voters). And then the post and some comments talk about UNS – historic voting patterns wiped out. All this without the slightest attempt of justification.

  29. No big surprise. Would be interested to see how the boundary changes would potential affect seats based on a regular Lib Dem turnout.

    This last result didn’t exactly convincingly show that coalitions are out as the LD’s only had to hang onto say 10 seats for that to be back on the table.

    Its more interesting to see the overall picture

  30. AW

    And, of course, these are on current boundaries. Any boundary review is likely to follow the usual pattern of reducing the number in seats in northern cities where there is a relative decline in population and increasing the number of seats in the south where the population is growing

    But isn’t it the case that urban areas (strong Labour) are growing, and rural areas growing more slowly? So won’t the boundary review actually help Labour?

  31. Just to demonstrate my point.

    Seat A: the Conservatives, against all expectations have a 9% lead, while labour increased its vote share. LibDems return from the Conservative barn (3%), rest of the LibDem remain tactical voting. Ex UKIP: 2% leave the Conservative barn and 3% vote Labour. So, suddenly Labour needs a 2% swing against the Conservatives. A huge task, but very different.

    I don’t say that it will happen, but projecting the current voting patterns into the future is highly questionable especially after the current elections.

    I saw that YouGov decided that everything was fine, but perhaps …

  32. To use Hawthorn’s approach: the target is the 25% who will be bothered to vote, but have no affiliation, while making sure that tactical voting takes place.

  33. I think the boundaries are drawn on registered voters. Many are unregistered in urban areas, although I feel they should still be counted, as presumably they also generate case work for an MP.

  34. Sorry, that was directed at Richard.

    I wonder whether it might be better to use census data when drawing boundaries?

  35. @ Cover Drive

    They are drawn on the basis of registration, but there are some false beliefs about registration (and about turnout) – post hoc really and hangover from 1992. I haven’t seen any evidence for it for the last 10 years that would significantly alter the situation.

    It would be nice to have it on the basis of population, but it won’t happen.

  36. @ Unicorn

    Your questions are right (I mean that), penetrating and perhaps a bit too sharp. Life is a bit too short too. I wouldn’t hold my breath for a cohesive answer.

    Over the last half a year or so you laid out your thinking and the evidence. I suppose that the learning from it wasn’t as much about the forecasting models, but the inputs to the models. These were erroneous, and judging from the responses from the industry we will never learn the causes of these systematic errors. You also had the trouble of not having access to the raw data, thus the errors couldn’t be corrected.

    You put a lot of effort into this. You refined your argument. You actually have a blueprint of analysis now, so you can step back.

    What is the question: the hypothesis (that has never been verbalised by anyone, not even by the polling companies), the VI of individual pollsters, the methodology by which these are resulted in (including the design), the forecasting model, etc?

    What do we do with the human bias, not only in the data, but in the analysis (I don’t want to guess your bias, but it must have been there)? Can it be overcome or be managed in a methodological way?

    What part of the analysis can be automated, and which part of it is about judgement, even if it’s an anathema? Relatedly, are we entitled to connect the individual points of the polls with a line? And if we do, what message does it give through our visual perceptions?

    When we do the analysis, should we follow our hypothesis (verbalised or not), or the way in which the data was created (e.g. that the U.S. Of parties must be 100%, but in reality we don’t add up party preferences, but allocate a certain percentage on the basis of the raw and processed data).

    How do we deal with magnitude? Yes, there is a statistically important effect – is it important. And reversely: this effect seems to be important, but statistically not significant.

    Is polling like building the most important part of a nuclear power station or building a shed at the bottom of the garden? Should I adjust the statistical arsenal?

    Unicorn, I mean all these with the greatest respect and appreciation to your contributions and discussions resulting from these. I also mean it in an extremely friendly way. I hope you continue your contribution to our understanding to the polls.


  37. Good Morning All, cloudy here in the Premier League Sea side town.

    Thank you, as ever, for your work. Labour in 1950 and 1951 also piled up support in ‘their’ areas, and lost votes to swing in other parts of E and W.

    I agree that voting patterns are volatile now, so even Scotland could swing away from SNP and England could swing away from Lab, enough to lose a majority.

    However, Labour needs a leader and an appropriate policy philosophy which will reach where voters are, not where she thinks they are. It should get rid of the mind-set of ‘ordinary’ folk.

    Off to school here I go.

  38. Interesting to see that tactical voting declined. I did predict this, along, if I remember correctly, with Bill Patrick. My explanation was that the expectation of a hung parliament with uncertain alliances and outcomes meant that a tactical vote might not end up with the outcome desired.

    Another smaller factor was that voters thought that ‘legitimacy’ might prove to be a factor, i.e. that a vote for Labour in a safe Tory seat still added something to Labour’s ‘right to govern’.

    I think Labour has every chance next time around, but it does require a policy mix and leader that will appeal to the centre ground. Its just so obvious, but I won’t ‘labour’ the point.

    I don’t see the EU referendum as a turning-point, and I anticipate that UKIP will decline. But a Tory civil war could develop, perhaps surrounding Cameron’s replacement, and there are always plenty of ‘events’. We are far from out of the woods economically, and the assumption of prolonged growth could be wishful thinking.

    As for the LibDems, it looks like a long time on the sidelines. I can see Farron becoming the ”Farage’ of the left. I think they could end up with a UKIP- style 12-14% of the vote but even fewer seats. But a bigger vote share might mean more council seats, which could be the way back.

  39. “Those factors have vanished.”

    I think sometimes we get a bit too statistically technical on here. Before this election, the system was skewed against Con, now – “Those factors have vanished.”

    So overnight, the sense of a pre determined bias for and against one side or another changes completely.

    All this proves is that those of us saying things like ‘Cons can’t win a majority due to their weaknesses in Scotland, northern cities, London, etc ‘ were completely wrong, as will be those who now say ‘Labour can’t win because….’. They can win, but whether they do depends on how people vote.

    In a FPTP system with built in tipping points and now four parties with significant vote shares and different distributions, any result is possible in changing circumstances.

  40. How the tables are turned. I remember all those posts telling us that Cons could never win again etc etc.

    ChrisLane, I think you are right to be concerned about leadership & a political philosophy with reach.

    I do not see a leader among the current hopefulls.
    Cameron can severely dent the reason to vote Labour if he really focuses on One Nation.

    But as this thread topic shows, nothing is forever in UK politics.

    UK Polling methodology is beginning to look as dated & clunky as the CD

  41. Does anybody have a decent spreadsheet with the results for 2015?


  42. This assumes no recovery to LD’s – from the lowest position and new leader the only way is up for them.

  43. Good morning all from a fresh and sunny East Renfrewshire.

    It doesn’t matter how big the mountain is for Labour.
    When they pick Chukka Umunna the ethnic block accompanied with white middle class guilt will be enough to see them over the finish line.

    And everyone will be like OMG first black PM, omg history, omg Obamania, omg bigger swing than Blair, omg Chukka so cool, omg Chukka so sexy…, David Cameron stale and pale
    Such is what politics is reduced to nowadays”

    You’re forgetting one thing..Boris!! He’s bubbly and cuddly and voters love teddy bears. ;-)

  44. @Andy JS

    Thank you for that spreadsheet. I had already found it from your twitter feed and used it to inform my comments on the previous thread, so this is a belated acknowledgement that I should have given then.


    I’m wondering whether you also used Andy’s spreadsheet for your analysis, since by yesterday I was unable to find a publically available comprehensive spreadsheet of all the election results, as opposed to one just focusing on Labour targets. If so, your analysis might be slightly pessimistic in terms of swing to become the largest party in seats, as the potential of the Conservatives losing seats to the LDs or possibly UKIP might have been overlooked, although we must be only talking about a handful of such seats.

    Another more substantive point is that any analysis which focuses only on the scale of a Con-Lab swing necessary for a Labour recovery is too one-dimensional. A Labour recovery in Scotland would have to be at the expense of the SNP, within what might yet prove to be a very volatile Scottish electorate.

    As such it overstates the scale of the mountain Labour would have to climb, a mountain though it undoubtably is.

  45. A lot will depend on which faction in Labour wins out. Will it be the, “we weren’t left wing enough”, (Diane Abbott, Trades Unions), or will it be the modernisers (mandelson, Reed, Etc)?

    One of that crowd will give labour a chance of election, the other will put them in the wilderness for another long period.

    Like Colin though, in the bunch on offer I don’t see any real leaders. Ummuna looks like he might be the most competent but it’s too early to tell.

  46. @MitM

    “the ethnic block”

    What planet are you on?

  47. For me the really disappointing aspect is the fact that Labour had a leadership that just did not connect with Scottish voters. If Miliband had been more appealing to core Labour voters during the referendum (like Gordon) then I don’t think the total collapse in Scotland would have been so extreme.

  48. @Santafemad

    Even Scottish Labour didn’t have a leadership which managed to connect with Scottish voters! Milliband didn’t even try. He said, a few days before the election that he ”got it”, regarding the Scottish situation. That was a complete fabrication – and everyone up here knew that.

    Murphy seems to have a good press in the south: “capable” “hard working” etc. but he wasn’t the man for the job. We knew that. The SNP were hardly able to contain their glee when JM was elected. And such is political life here in Scotland that no-one who is outside Holyrood can now hope to engage.

    And, lastly, the head of the party in Holyrood has to be in charge of the MPs as well as MSPs. Until Labour ‘get’ that, then LiS will struggle.

  49. The Tories are in because Labour fouled up (before and after 2010).
    Labour will be in when (not if) the Tories foul up, unless Labour continue the foul up of neglecting their core support and UKIP take it.
    The other possibility is that in response to UK pressure to ‘change Europe’ the EU adopts measures and arguments which make membership unattractive to voters. UKIP support will rise when Cameron announces cosmetic changes as major EU reform. That rise will come at the expense of a Labour party which would deny a referendum.

1 2 3 15