ICM’s monthly poll for the Guardian is out. Topline voting intention figures with changes from ICM’s last poll in January are CON 35%(nc), LAB 38%(-1), LDEM 18%(+3). The three point Labour lead is very close to Populus’s (who use a similar method to ICM’s), and to YouGov’s four point lead at the weekend (though I still expect that one was a bit of an outlier). The reallocation of don’t knows in Populus and ICM’s polls should be expected to produce slightly lower Labour leads than other pollsters.

More surprising is that 18% for the Liberal Democrats, their highest score since October. We’ve seen quite a lot of variation in the levels of support different companies give the Liberal Democrats this Parliament, but recently it had looked as though they were starting to come together. Looking at all the companies’s most recent polls YouGov have them at 10%, ComRes, Populus and Angus Reid at 11% and MORI at 13% – there’s no obvious explanation for the large contrast with ICM’s figure.

ICM also have some interesting findings on the AV referendum. They have voting intention in the referendum standing at YES 37%(-7), NO 37%(-1), Don’t know 27%(+9). Changes are from the last time ICM asked the question in December – this is the first time that a poll that was not prompted with explanations of the system or pro- and anti- arguments has shown the NO campaign catching the YES campaign. Note that the fieldwork was conducted between Friday and Sunday, the first since Cameron’s and Clegg’s repective pro and anti speeches and the first since the media have starting to devote some attention to the campaign.

UPDATE: YouGov’s daily poll for the Sun has topline figures of CON 36%, LAB 42%, LDEM 11%. After 7 polls in a row with the Labour lead at 7 points or above, including two ten point leads, we’ve now just seen three YouGov polls in a row showing the Labour lead of below 7 points.

96 Responses to “ICM show Lib Dems up and AV and FPTP neck and neck”

1 2
  1. Combined Tory and Labour tally of 73% with Labour 3% ahead at 38% feels about right to me, but the Lib Dem rating of 18% looks to be on the wilder fringes of MOE, I would think. If it’s true, however, and we’re seeing a revival in Lib Dem support, what do we think might be generating it? There’s no noticeable drift from either Labour or Tory, so it’s either DKs opting for them or some strange churning going on amongst others. I follow political events quite closely and I haven’t seen any recent development that would explain a Lib Dem surge in popularity, unless there’s been another Leaders Debate on TV I might have missed!!

  2. That LD result looks like a real outlier does it not? However, strangely, given that if a disproportionately large group of LD voters was stumbled upon, the AV shift from YES to DK looks weirdly out of sync with this? That is unless AV is massiveley unpopular as soon as the workings of it become apparent, even amongst LD type voters!
    To me it all looks very strange?
    What does everyone else make of it?

  3. The list of poll results on this site suggests that the last time the LDs hit 18% was 7 October last year (ICM).

    Good for LD morale, if nothing else.

    The AV referendum poll is less welcome…

  4. Increased LD support due to their additional public appearances/press coverage in support of yes to AV campaign ?

    Is this possible and if so will this come out in other polling ?

  5. Doesn’t say a lot for the quality of our media. They start covering the issues involved with AV and there’s a big spike in the number of “don’t knows”!

    I have to say, after the Feast of YouGov these intermittent polls from other companies do rather feel like Famine. How are we supposed to know if a poll is an outlier when it’s once a month..

  6. Last time yellows hit 18% with ICm YG hit 12%.. so if we see an uptick in YG we’ll know that the ICM is reasonably ‘on song’… who is right at this stage in the polling cycle is less important, I think, it’s all about trends…

    In the five years pre-May 2010 ICM/YG disagreed over yellow to the tune of 2.5% anyways so I wouldn’t get yor knickers in a twist.. as we say with YG? CIm when yellow vtoe really depreciates they all come tumbling down.. that ICm is expanding again is confirmation for me that yellow are back.

    The day the red srategy switched to C led G was Ed’s equivalent to operation barabarossa…

    He’ll live to regret it.

  7. We know that ICM tends to give the LDs the highest poll ratings.

    In April 2010 the Leaders’ debates produced an artificial spike in LD poll ratings that turned out not to reflect real voting intentions.

    I think what we may be seeing is that the opprobrium heaped on the LDs in Nov-Dec over student fees produced an temporary depression in LD poll-ratings that is now unwinding – because students fees are no longer the hot topic of the day.

    I don’t think we’ll see 18% from other pollsters, but I’m hopeful we might begin to see figures closer to 15% than 10%.

  8. Anthony

    Do we know any details of this poll. For example the Party voting intention of the 7% who have shifted from YES to DK?
    On second thoughts I wonder if the LD increase is because Clegg is seen to be standing up to Cameron on something, rather than the “Nick’n David lovefest” we have had since last May? Perhaps the Labour YES people’s unwillingness to be associated with the LD YES people has led Labour voters to weaken in their favour of the YES option. Perhaps some Tory voters were previously lukewarm YES, thinking that it was an agree compromise of the Coaltion, not fully realising Cameron was to so strongly advise a NO – so they’ve switched?
    Without the breakdown of the figures this is all guesswork!

  9. @Eoin

    Agree. Would be very happy with a 12% from YG.

    I thought ‘Conservative-led Government’ was meant to be a subtle attack on the LDs – ‘they’re so irrelevant they’re not worth mentioning’. Would be a delicious irony if it had led to increased support for the LDs.

  10. @Eoin
    Agree. Would be very happy with a 12% from YG.

    I thought ‘C led G’ was meant to be a subtle attack on the LDs – ‘they’re so irrelevant they’re not worth mentioning’. Would be a delicious irony if it had led to increased support for the LDs.

  11. Much as i’d like to believe the 18% showing for the LD’s, we need to consider that last month ICM also had the LD’s at an improbable 15%.

    However the margin between the big two seems about right at approximately 5% and falling. I expect YG to confirm this tonight.

    Part of the advantage of being in Government is that you get to control the agenda, and we have seen some deft use of this by the Tories and the LD’s in recent days.

  12. Even as an LD I would say this looks like the outside of MoE.

    Maybe Clegg should make a point of openly disagreeing with Cameron more often. Being seen on the opposing side seems to help the party.

    On top of that, there haven’t been any major stories in the press slamming the party or key leadership figures – which makes a change from the last, ..errr… forever.

  13. D ab,

    Now the C led G stuff was meant to give the yellows respite and turn the guns on blue.. in fairness to Ed it acheived an accelerated decline in blue.. but at what expense?

    Pugilist meta[hors spring to mind.. “a rumble in the jungle round8 perhaps?”

  14. This is probably a good poll for the LDs, and a bad one for Labor, because this poll seems to confirm the narrowing Labour Lead.
    As for the conservatives: it can get better, as the latest polls show, but there’s still some work to do to get back in the lead.
    Maybe this won’t happen before 2013, but you never know!

  15. This poll is good for the Liberal Democrats and bad for the Labour Party. I expect the conservatves to recover a bit, my prediction for the next poll of Yougov is:

    Labour: 41-42%
    Conservative: 37%
    Liberal Democrat: 9-11%

  16. How many say they will bother to vote on the issue?

    I suspect many of the I don’t understand / cant be bothered / all change is bad lot wont get out of bed for it.

    The many who care about issues and minor parties will vote.

  17. In YouGov, we’ve seen a slight increase in the Lib Dem vote to around 10% since the beginning of the month which has been coupled with a bigger drop in support for the government from Lib Dem voters. This suggests some signs of dissent from the Lib Dems in government is popular with their disaffected voters.

    Now ICM’s weighting methods tend to exaggerate the movements we see in YouGov, both because of their reallocation of past voters and more subtly because of the weighting they give to likelihood to vote. This helps explain last month’s ICM 15% and and probably the increase to 18% as well. A similar movement to YouGov’s is reflected as May 2010 Lib Dems become a bit less disillusioned and increase their likelihood to vote.

    However, unless Lib Dem dissent is seen to be effective in altering the government’s policies, that support will either ebb away or be used to bring pressure for internal change in the Lib Dems. We won’t see anything before May now, but whatever the results then, there will be demands for changes afterwards. Or we’ll turn on al-Jazeera one day and see massed crowds in Cowley Street calling for the overthrow of the corrupt capitalists lackeys and a change in leadership.

  18. Latest YG

    Lab: 42
    Con: 36

    App: -24

    LDs creeping up.

  19. all very strange, and almost certainly an out liner for the LDs. As for the AV vote, it seems it’s all down to the don’t knows. I’m still of the mind that as the day gets closer and people become genuinely interested, some of the “purple” posse will be telling politicians they want change and to not support them would be a curse on their children.

    in other words, labour will reluctantly give their thumbs up for their supporters to vote “yes”.

  20. @ Alan Raby

    Excuse my ignorance, but who are the “purple” posse?

  21. Local government elections are about 10 weeks away.

    LD’s are known for their skill in fighting these. Would I be stupid to suggest that could be having some effect on the VI polls?

  22. There does seem to have been a slight narrowing of the Labour lead in the past week – although I would contend that tonight’s 6-point margin does suggest that Sunday’s YouGov poll was a bit of an outliner.

    By the way, Anthony, could you please put up the tables for the Sunday Times/YouGov poll (or have you done so – because I seem to have missed them?). The weekly tables are essential reading for those of us who want to study the entrails of the poll.

    Regarding AV, what’s the reason for the drop in support for “Yes”? Could it be the launch of the “No” campaign – and their highly effective marketing tool relating to the alleged cost of the proposed voting system?

    Or might it simply be that the other polls are converging with the findings of the YouGov referendum poll (thus effectively endorsing the latter’s decision to begin with an explanation of AV?).

    Whatever the reason, I would still contend that YouGov should now ditch the opening explanation – and start a new trend line based upon a simple “Will you vote Yes or No in the forthcoming AV referendum?” question.

    People are becoming better informed about this issue – so an artificial lead-in is simply not necessary (and could potentially cause major inaccuracy).

  23. @Tony – it’s a movement of people who basically want fairer voting (PR) but will be campaigning for AV. They’ve been a little silent so far, but will be quite vocal closer the day.

    You will occasionally hear references to people wearing purple, which is linked to them.

    I think the website is takebackparliament or something like that.

  24. Amberstar has said exactly what I was going to say – Lib Dems are already on the streets knocking on doors and people are expressing their choice for the local elections

  25. @ Alan Raby

    Thanks Alan – I’ll look out for them!

  26. A bit of Con-led ecovery is probably to be expected seeing as how the news has mainly been full of AV (with Cameron and Clegg being given lots of uncritical coverage) and Libya/Bahrain, with a little bit of workshy scrounger bashing thrown in. For the moment, the feeling of impending doom has perhaps been lifted a bit. I rather doubt that it will last.

    Although the long-term trend has been steadily upwards for Labour, there have nevertheless been some temporary blips. There was one at the start of December, and another at the end of January. Each blip lasted no more than a week before normal service was resumed. I think we’ll need to wait at least another week before we can decide whether this is another blip or something more substantial.

  27. There has actually been some more measured coverage of Lib Dem influences in the news rather than the hyperbole we saw up to Christmas and into the early part of the year.

    Perhaps the message is finally getting through that the Lib Dems are having some influence.

    Let’s hope so, because they ARE!

    Not too worried about the AV result…i still think the momentum is with YES especially given the oddities supporting NO

  28. Reckon the Economy on the ground has improved a little.
    Expect decent numbers for growth in Q1&2 and unemployment increase to be lower than Q4 ’10. Inflation which we are all feeling is the main drag on Gov’t support Economically speaking.

    Q3 2011-Q2 2012 is a curucal period Ecomically as this period will self evidently reflect the Gov’ts policies and any good or bad news can not be assigned to the last lot.

  29. Thanks, Hunty :-) I don’t feel so stupid for suggesting it, now.

  30. If the situation in the Middle East gets worse, we could see a big increase in fuel prices. The government would then be put under massive pressure to reduce fuel taxes, if prices start to hit £1.40 a litre of more. That would put the price of filling a Mondeo to about a £100.

    I would not like to be in George Osbornes shoes if this were to happen and he could not reduce the fuel taxes, even if the government were receiving more revenues from North Sea oil.

    This would hurt the coalition parties, even if they were not to blame for the situation.

  31. @Amber Star – Your comments can never be characterised as lacking in good sense imo.

    For a LD recovery though, it’s got to be Stockholm syndrome. :)

  32. I think the U-turn on forests has been good for the governing coalition; people – rightly or wrongly – believe that protesting/ pressuring the government will head off anything that they are strongly against.

    It will be interesting to see what happens if (when?) the government don’t u-turn on something that is important to floating voters.

  33. @ Billy Bob

    Thank you :-)

  34. Amber

    “something that is important to floating voters”

    lifebelt provision?

  35. @Oldnat,

    Surely lifebelt provision is only an issue for sinking voters.

    Floating voters may well take an “I’m all right, Jack” attitude to bouyancy, and therefore vote Tory….

  36. @R-Huckle – I suspect they could reduce the tax a little in that circumstance, as they would be receiving quite a bit extra by then. Effectively they could get one of their manifesto pledges in because of it.

  37. Amber

    If there was a local election effect for the Lib Dems due to canvassing etc we’d probably see a Feb to May rise in other, non general election years. On very quick examination there seems no pattern. Of course if you live in one of the more goldy-coloured bits of Edinburgh in the run up to an SP election, it may feel like that. ;)

    Robin Hood

    Tables for Sunday Times went up this morning:


    and to compensate for lateness we’ve even go tonight’s poll up early on the archive along with the latest AV poll and a very interesting poll of Egyptians from YouGov Siraj.

  38. Neil A


  39. Quickly going through the details of tonight’s poll here:


    you get the impression that things have not perhaps got better, but certainly not got worse for the coalition over the last fortnight. The economic trackers (now with added ‘necessary’, ‘shallow’ and ‘slowly’ – thank you Anthony) are all the same or maybe a point better and the blame for the cuts has shifted back with an increase of 3 points for Labour.

    Most interestingly all the ‘worry’ indicators – whether people fear that they will suffer from crime, bad health, job loss, foreign competition etc – are all down, some by a significant amount (YouGov samples sizes have gone again this month to around 2,500). I’m always a bit sceptical about these, partly because Brits love to moan about anything and also because some of the questions are about things the government can’t do much about.

    I mean the answer to Thinking about the next two or three years, how worried are you that people like you will suffer ill-health has little to do with the government – as opposed to getting the right/speedy/any treatment for it. No one is suggesting that Lansley will re-release smallpox into the community to thin out the poors or Pickles will invent a vaccine to prevent hemorrhoids.

    None the less I suppose the answers to such questions give an indication of the general feeling of well-being towards the future and as such they have improved a little, though still patriotically pessimistic.

    The only slight consolation for Labour is that a Labour government led by Ed Miliband has finally drawn equal with the Conservatives as a favourite for after the next election. his figures were down in the Sunday Times, but tonight’s figures suggest that that particular poll was a bit of an outlier.

    Things that only I am sad enough to notice part 94 (apart from the tables still missing from last week): the weighting pages are looking very odd.

  40. @ Roger Mexico

    Thanks for the link.

    The 18-24 break does look a bit quirky: it gives the Tories an 8-point lead over Labour, which is completely out of line with recent trends among this group, and possibly explains why this poll may be a bit of an outliner.

    [Have I done my sums correctly here? The 18-24 sample size is 297, so an 8% lead among this sub-set would represent c.24 respondents and contribute over 1 percentage point to the overall Tory poll lead – whereas normally the group leans heavily towards Labour].

    When I worked at MORI in the 1980s, we periodically combined our polls into a quarterly “aggregate analysis”. This smoothed out the variations and enabled us to analyse with much greater confidence how different demographic sub-sets were behaving. (The client for our individual voting intention polls at that time was The Times – but I seem to remember we also sold them the findings of the aggregate analysis too, so good was the quality of data).

    Any chance of YouGov doing the same thing? It could be particularly informative in understanding the new structure of the Lib Dem vote – something which is difficult to ascertain from an individual poll because of the smaller sample size resulting from their reduced share of the vote.

    In the 1980s, the aggregate analysis enabled us to identify that there was a strong skew towards the SDP-Liberal Alliance among ABC1 Trade Unionists (i.e. white collar union members). As an SDP activist at that time, this sort of finding was of particular interest both to me and to Peter Hutton – the sole other social democrat among MORI’s then political team.

  41. Robin Hood

    Actually when you take into account the 28% non voters and the very heavy weighting up to get the 18-24s to their proportion of the population from a sample of only 130, an 8% lead in that group could be only 7 or 8 people. So 4 people saying Con rather than Lab gives you 1 point change (I know it’s not that simple but it’s that order of magnitude).

    That segment of the population seems particularly vulnerable to wild swings because of the small number in the samples, especially in the ST figures. Con/Lab/LD was 23/54/10 the previous Sunday and 30/51/7 the week before. It was also a Sunday Times published poll that lead to this stupidity:


    I reckon that they all come home hammered on cheap students’ Thursday night drink deals, type any old rubbish into YouGov and forget about it. Late Friday afternoon Anthony looks at it, despairs, and heads of the pub himself.

  42. This is a bad poll for supporters of AV (though I caveat that think most of the polls have been bad for them). This poll is one that shows the Lib Dems the highest they’ve been in a long while, tends to show a higher Lib Dem percentage than most, and here shows an upward trend for Lib Dems. My feeling is, at least between the voters of the three major parties (and I don’t know how Scottish Nats, Welsh Nats, or Greens would vote), most Lib Dems will vote yes on AV, most Tories will vote against AV, and Labour voters are the swing voters who likely would decide the election result (and it’s not clear which way this vote will go). In this context, the fact that an increase in the Lib Dem vote, the core vote for yes, has not increased the number in support of AV is problematic for the yes side.

    I still think that measuring a referendum vote needs to be analyzed to take into account the percentage of yes voters, not the lead over no that yes might have. A referendum vote is not the same as a traditional election and the polling is usually not the same. Thus, I still stand by my contention a 37-37 polling tie is not really a horserace but really shows the Yes vote trailing.

  43. @ Nick Hadley/Crossbat11

    Most of the polls we’ve seen show very minor fluctuations between the parties. Therefore a 1% drop or gain here or there doesn’t mean that much. Compared to other pollsters showing larger leads for Labour, I don’t think this one can be considered a significant drop for Labour because the last poll only had them up 4%. It’s also possible that this poll may simply be one that tends to state Lib Dem support as higher than it typically is.

  44. I’m expecting the mass privatisation agenda coupled with economic stats between now and end of April to put Labour miles ahead.

    Nobody voted to privatise state provision, few want it and it will lead to a meltdown in the Government vote.

    If Con vote falls dramatically and Lib Dem recovers a smidgeon maybe Lib Dem will see that pulling out will improve their popularity. Might even get those 2nd place Lab votes.

    Privatisation is very unpopular now. People have seen where the “efficiencies” of the Market lead. Polarisation of wealth and eventually economic meltdown.

  45. @Roger Mexico – “… type any old rubbish into YouGov and forget about it.”

    Aye (as Mike N might say), and not just the students!

    What proportion of respondents ‘spoil’ their questionaires or for other reasons need to be filtered out?

  46. Robin Hood – “Any chance of YouGov doing the same thing?”

    We’ve done it before, and no doubt we’ll do it again. A little patience :)

  47. At present no pollster has yellows below 11%.

    given that they were 7s and 8s for large chunks of Dec/Jan this is in relative terms quite the recovery.

    If the Ipsos Mori Soctland poll is to be believed, then some of that recovery is happening in red heartlands.

    I am wary of that conclusion myself but it is right to point out where the stats are leading us… Conflcitingly, YG’s London poll last week still showed the Yellows down and out there…

    There is no reliable data to show it but I wonder if some of the dip of blue from say 40 to 36 has meant that they are moving over to the yellow camp.

    Needless to say, I am looking forward to the YG Wales poll when it emerges.. if it shows a 1-2% jump for yellows in will be quite instructive.

    a yellow gain in the red heartlands ‘might’ make it difficult for reds in the regional assembly elections.

  48. Looks like the end of the beginning of this government to me – Lab have climbed steadily but now appear to be flatlining at low 40s (in YG polls), Cons have settled around their GE score of 36-37, LDs have dead-cat bounced from 7-8 to around 11.

    People seem to have got used to the major/minor partner roles of Con/LD in the coalition govt, and got a bit of an idea about what Ed is (not) doing. The May elections will add a bit of turbulence to the polling data, but I predict that, say, 42/37/11 will be the steady state for the next while.

    Game on.

  49. Crying wolf only works a few times as Labour are starting to find out.

    Since the GE they have been predicting the end of the world (or at least the socialist eutopia they deisre) as we know it with the prospect of mass unemployment and mass suffering.

    Of course this has not happened because the Govt is still spending billions of borrowed money to prop up the bloated state sector. (as per fat cat council bosses and non jobs)

    People are finding out that for most of them life still goes on just fine. They may have to make some adjustments to their life styles, less going out (but we drink too much anyway) less on the food buget ( but as a nation we are suffering from obesity) and a few less recreational miles in the car (good for the environment).

    However overall most people are not suffering, and so are beginning to hear Labour doom and gloom predictions as just a background noise and blank it out.

    IMO Labour has peaked. and as spring and the royal wedding approach the feel good factor will come back and I expect the parties to be neck and neck in the polls again by mid year.

  50. @ John Fletcher

    “Of course this has not happened because the Govt is still spending billions of borrowed money to prop up the bloated state sector”

    That’s a bit of a funny one actually. Before the Q4 figures Osborne was boasting about how they’d cut back on spending and it wasn’t having an effect on GDP. After the Q4 figures he was saying that the gov’t had been spending more than it had in all previous years. Weird eh?

    (As for the public sector – I’m sure a healthy amount of that money has being going towards banks so that they can still pay their ceo’s their bonuses)

1 2