Through December there were some somewhat contradictory polls – we saw YouGov putting the Conservatives way up at 45%, then an ICM poll showing Labour recovering. That was followed by a YouGov poll that also showed Labour recovering…but that was taken in the few days before Christmas when it’s had to believe a reliable sample could have been drawn.

It looked like Labour might have been recovering. Populus’s poll this month didn’t show the same sort of recovery in Labour’s support, but it did show the Tory lead falling thanks to them loosing support to the Liberal Democrats. Now two new weekend polls, one from Ipsos MORI for the Sun and one by YouGov for the Sunday Times suggest the Conservative lead is back up into double figures.

MORI’s topline figures, with changes from last month, are CON 42%(nc), LAB 32%(-3), LD 15%(+1). YouGov’s figures are CON 43%(+3), LAB 33%(-2), LDEM 14%(-1) – though that is comparing things to the poll done at Christmastime, comparing it to the previous YouGov poll the Conservatives are unchanged, Labour up 2, the Lib Dems down 2.

We can’t tell exactly what happened – it could have been that ICM’s poll and the Christmas YouGov polls were just blips or artefacts of the seasonal timing, and that actually the picture is pretty stable with the Tory party stable at around 40%, Labour recovering ever so slightly but still in the low thirties, Alternatively it could be that Labour had been recovering, but have been put back in their box thanks to the Hain funding row which has been ticking over during the week – both MORI and YouGov’s fieldwork was done mid-week. It’s now pretty irrelevant, whether there was a recovery or not, these two polls suggest Labour are back down in the low thirties.

On the Lib Dem front, there’s a contrast here between the Populus poll which showed them three points up and the MORI and YouGov polls which show no Clegg boost.

UPDATE: There is also a new ICM poll in the Sunday Telegraph. The topline figures with changes from the December ICM poll are CON 40%(+1), LAB 33%(-1), LDEM 18% (nc), so again the Tories up around 40%, Labour down in the low thirties but recovering slightly (remember the changes here are comparing the poll to the ICM figures that showed a big increase last month – 33% is still an improvement on Labour’s November’s figures) and no obvious boost for the Liberal Democrats.

UPDATE 2: Some more interesting stuff in the polls – the full tables of the YouGov poll are up here, I’ll have a proper look tomorrow.

90 Responses to “Tories back to 10 points ahead”

1 2
  1. I just want to check Mark that you are still sticking to your autumn forecast of a great Lib Dem recovery in the opinion polls by the Spring? Including YouGov.Isolated council bye election gains ie as in Thatcham last week don’t count but as I said before the May elections do as they will be measurable on a wide front against previous years. Are you also Mark forecasting that the Lib Dems will-as in 2006- outpoll Labour in the May elections? And do you think that any Lib Dem losses to the Tories will be less than last year,about the same or worse. Intrigued as to your answer.

  2. Nick
    Yes my forecast is that LibDem poll figures May time will be around the same level as they were last May .
    My forecast for the May locals is entered in the competition , Labour to have a net small number of gains circa 70 with Conservatives and LibDems both losing a small number of seats 20-30 each . Labour will be able to spin this as a success but in reality they will be bumping along the bottom as all the seats being fought ( only 1/5th of those fought last year ) and were last fought in 2004 which was a very bad year for Labour anyway . As the majority of seats being fought are in the Met districts , Wales and other urban areas , Labour may actually outpoll the Conservatives as well as LibDems in actual results but not of course in notional figures that allow for the fact most of the Conservative leaning districts are not voting .
    LibDems v Conservatives gains/losses should be virtually neutral as in deed they were last May in these districts as opposed to those areas which had allout elections last fought in 2003 .

  3. Anthony , many thanks for the detailed reply , I will respond further when I have time but having to do some proper work at the moment .

  4. Anthony,


    I am aware of the severe limitations of the figures for Scotland, but they are all I’ve got.

    As such I try to look for definite trends If I think I’ve found them.

    That’s why 7% for the LibDems looks unusual, it could just be an error but it looks a bit large for that. With the best periods for the LibDems in over a decade coinciding with two Scots at the helm, I don’t think it’s unreasonable to suggest that not having a Scottish leader might well have had an impact.

    I can think of three reasons.

    1) That could particularly be the case because of lack of recognition, Cambpell and Kennedy were household names (literally…) in Scotland before becoming leader, Clegg is an unknown.

    2) Clegg seems to be to the right of either Kennedy or Campbell and a bit like a tort, not a great asset in Scotland.

    3) Rightly or wrongly many Scots feel that both Scots leaders were stabbed in the back, and that hasn’t helped the Libdems image, particularly the “We’re the nice guys, you can trust” bit.


  5. “LibDems v Conservatives gains/losses should be virtually neutral as in deed they were last May in these districts as opposed to those areas which had allout elections last fought in 2003 .”

    I think this may be an attempt to spin the LD ratings – which may be more favourable against the Tories in some of the areas voting this year,
    but is deeply misleading nevertheless, because the
    overall picture in May 2007 was the LDs taking a hammering
    by the Tories (and sometimes even Labour).

  6. Joe , sorry it is not spin but fact . As an example for the Met districts all last fought in 2004 ( as they will be this year ( 1/3rd of all the seats being fought this year ) , the actual gains/losses were :-
    Con 19 gains from Lab 9 losses
    Con 3 gains from LibDems 10 losses
    Con 2 gains from Others 1 loss
    Lab 22 gains from LibDem 16 losses
    Lab 5 gains from Others 2 losses
    LibDem 2 gains from Others 1 loss

    Net Con +4 Lab -1 LibDem +2 Others -5
    I am not interested in spinning what happened last May that’s history , but giving reasons for my forecast as to what will happen this May .

  7. Mark
    Actually the Lib Dem figures last May were not that hot with the average a little over 18% as compared to a bit over 16% today so you are only forecasting an additionally 2% which seems somewhat modest.I think Mark it has to exceed that before you can lay claim to a significant revival. Nevertheless fair enough you have posted a forecast so lets hear from Mike Richardson (Con) and Steve Wheeler (Lab) as to where percentagewise they think their parties will be -on average-in the polls come May.

  8. I would hope (as a Tory) that in share of vote, we exceed the 40/41 of May 2007 (in 2004 it was 38%).
    I am prepared, for some Labour gains (maybe gains and losses on both sides) because this is a fairly small round of elections (unlike 2007) and not our most fertile areas, although we need to show improvements in them too.
    Labour should do better than their awful 26% in 2004, and probably above the 27% of 2007.
    If they are above 30 per cent, then there should be some Labour gains.
    Perhaps the Lib Dems will go down with about half each going to the other 2.

  9. We are voting in my own area on Feb 7th.

    In fact we have 2 votes that day – one is a local council by-election and the other is a County Council election.

    I agree with previous views that by-election result should be treated with a very large pinch of salt but they are real votes and perhaps some useful information may come out of the result – such as a higher than usual turnout or a large swing to one particular party.

  10. Joe , guess you are talking the notional vote shares rather than the actual vote shares . For interest these were :-
    Con 2004 25.9% 2006 27.0% 2007 27.0%
    Lab 2004 32.6% 2006 34.0% 2007 34.4%
    LD. 2004 25.5% 2006 23.6% 2007 22.0%
    Con 2004 30.9% 2006 33.5% 2007 32.9%
    Lab 2004 29.6% 2006 28.5% 2007 28.7%
    LD. 2004 24.1% 2006 25.1% 2007 23.8%
    Con 2004 41.7% 2006 44.6% 2007 44.3%
    Lab 2004 21.8% 2006 20.1% 2007 19.6%
    LD. 2004 25.3% 2006 25.3% 2007 23.7%
    Wards won
    Con 2004 1012 2006 1015 2007 949
    Lab 2004 .784 2006 .743 2007 712
    LD. 2004 .604 2006 .579 2007 553

  11. Anthony , thanks for your detailed response , some of which I do agree with . Yes weighting can and probably does make full amends for a bias in the sampling for demographic factors such as age/sex/ABC’s . No I do not know the make up of the Yougov panel but there is much anecdotal evidence that in the last 2/3 years an increasing number of politically active people have joined myself included and for example the urgings of Iain Lindlay on his site . Has it been enough to make the panel unrepresentative of the population as a whole a la Readers Digest in the US in the 1940’s – I believe so .
    For evidence , I quote both the unweighted and weighted figures from the latest polls
    Y/G unweighted Con 35.7% Lab 24.9% LD 9.6%
    Pop unweighted Con 24.2% Lab 20.7% LD 10.2%
    ICM unweighted Con 25.6% Lab 21.7% LD 9.2%
    Mor unweighted Con 27.8% Lab 22.3% LD 10.1%

    Y/G weighted Con 33.8% Lab 25.8% LD 10.6%
    Pop weighted Con 23.0% Lab 19.9% LD 12.0%
    ICM weighted Con 24.3% Lab 21.0% LD 11.0%
    Mor weighted Con 27.5% Lab 21.9% LD 10.2%

    Clearly the Yougov figures bear no similarity to those of any other pollster either unweighted or weighted and the difference is not simply one of being related to Yougov getting voting intentions from a higher 5 of respondents than other pollsters . Yougov are getting a greater % of responses saying they will vote Conservative and a lower % saying they will vote LibDem than all the other pollsters including Mori .
    This discrepancy is before we get into whether weighting for past vote is justified or not and to what level and it is right that someone should question whether it is .
    Does it all matter that we have this discrepancy . I feel it does . Prior to the last election all the pollsters were in pretty close agreement with their findings , now they are not . Variations from poll to poll due to ampling errors are to be expected but to me it is worrying that such a large difference has developed . Now it may be that Yougov are correct and the other pollsters are wrong but they cannot all be right and in something which is or should be a science this should not be the case .

  12. Thanks to Mark for his detailed response, and for those figures which I found useful.

    As he well knows,
    the BBC, and Thrasher and Rawlings do a national projection from a large number of key wards, because they are not all contested each year, although 2007 was of course a big round, but with nothing in London.

    There are also difficulties doing projections from places where there are lots of Independents.

    Last year it was 40/41 to 27.

  13. Joe , I should have made it clear that the figures I gave above for 2007 are for the Unitaries and Districts with only 1/3rd elections and do not include the substantially higher number with all out elections . They are therefore pretty much comparable to 2004 and 2006 .

  14. Leaving that that anecdotally you think that lots of political people have joined. I can quite understand why you’d think it, in the public eye YouGov are associated with political polls so you’d expect people interested in politics to join. It doesn’t actually work like that. Most panelists don’t join through the website, they are recruited through other surveys or projects that are non-political. If panellists are recruited through a survey about pizza, why would they be unduly politically minded? I know all this, and see the break downs of the panel, but it even surprises me to be honest. I remember when we were doing some polling of party members and I looked at the actual numbers of party members we had amongst panellists, I expected a greater proportion of YouGov members to be party members than amongst the population as a whole and was actually quite surprised to find that they weren’t vastly over-represented.

    The unweighted figures have no reason to be the same across different sampling methods (there is no reason why they should be. Phone polling for example tends to produce raw samples that contain too many Labour supporters, no one is sure why. It doesn’t matter because the phone pollsters correct for it). You’re left with weighted figures.

    There are a number of potential causes of the differences, and it’s probably a mix of lots of them. One is YouGov doesn’t get refusals as part of the sample. Phone polls tend to get about 7% or so who refuse and I guess people who refuse to answer questions probably don’t join an online panel in the first place.

    Secondly the phone polls tend to get a much higher level of don’t knows. Within that there are different potential reasons – it could be that the samples contain different proportions of people who don’t know if online panels do have more committed and political people, on the other hand it could be an interviewer effect: as you will know, ICM and Populus believe that many of those people who say don’t know are actually just reluctant to give their real voting intention to an interviewer and reallocate the responses accordingly. It could be a bit of both, or something else entirely.

    Thirdly there is the weighting the different pollsters use – we know it makes a difference in support for parties, but they also have to decide what levels do they weight people who didn’t vote in 2005, or don’t remember their vote in 2005 or say they don’t idenfify with any political party? They differ, and therefore lead to different proportions of people not giving a voting intention. Some pollsters could be weighting them wrongly – hell, they could all be weighting them wrongly.

    The blunt truth is that the raw samples don’t even pretend to be random at all in the case of MORI and YouGov, and aren’t actually random in the case of ICM and Populus. No one pretends they are representative when they are unweighted. When they are weighted, what they are weighted to is entirely the choice of the pollsters and they can be wrong or right, there is nothing to judge them against and consensus does not necessarily point to who is correct, as Harris showed in 1992 and ICM showed in 1997.

    Political polling would be a science if everyone could get genuinely random samples with 100% response rates and get respondents to always give accurate and truthful answers. None of those things are the case, in the case of the first two, pollsters don’t even get close, not in the same ballpark. In many ways it is more an art than a science.

    In the final published figures polls sometimes get different answers because they are actually asking slightly different things in terms of turnout or dealing with don’t knows. There are other differences because of sampling differences, weighting differences and so on. Until the mythical day comes when polls are perfectly random surveys with 100% response rates and cannot possibly be improved, I think different results are good. That’s the way things improve, some companies get the wrong, other people get it right, people learn, methods improve.

  15. The bottom line in all of this however is that the YouGov C Lead comes out on average as the same as that of all the other pollsters aggregated, just slightly less variable. “There are nine and sixty ways of constructing tribal lays, and every single one of them is right” (great quote I recently found – see here if u r interested)

  16. At least this site (unlike others) does not seem to go overboard in paying undue attention to the latest poll. A friend asked me recently what I thought the outcome of the next general election would be and I said we have almost no evidence at all as yet.
    If pressed I’d hazard that there’s about a 30% chance of another Labour overall majority, 55% chance of a hung parliament (evenly between Lab and C largest party) and 15% C overall majority – and that almost whatever happens in 2008 I’d be likely to say much the same at the end of the year, unless an election were to be imminent – and with an equal lack of confidence that this would be right!
    On turnout, Joe: the factors that cause high turnouts are usually ideological disparity (no), a close result (likely) and a national crisis (eg 9/11 in USA)(unknown)- so that between them it looks as if it matters whether one votes.
    As at least one of these factors is likely to apply, I’d expect a somewhat higher turnout than last time, but with the general trend downwards it may well not exceed 65%.

  17. Welcome to this site Robert Waller-we are for the most part friendly to each other here and reasonably respectful of each others views . Occasionally Anthony has to indulge in some censorship to keep us on the straight and narrow but not often.
    Are you the same Robert Waller who as a leading pundit of his day predicted a big majority for Harold Wilson over Ted Heath in 1970 and continually dismissed Margaret Thatcher’s chances or was that someone else? I see that you only rate the Tory chances of gaining a majority at the next GE as 15%!!! Why so high? That would be unlike the Robert Waller I am thinking of. Mind you is there any chance of me having a wee wager with you at those odds? In these days of credit crunching and worthless Northern Rock shares a man must seek more reliable sources of income wherever he can.

  18. Nick Keene
    I notice you say ‘for the most part’ friendly and ‘reasonably’ respectful before trying to have a (doubly inaccurate)go at me!
    1. I wouldn’t have done my GCSEs (or O levels) in 1970, so it would have been both flattering and foolish to laud me as a pundit then …
    2. On my hesitant and under-trailed opinions, which so seem to offend you, did you bother to get through to the bit where I said “with an equal lack of confidence that this would be right” ?
    This would have suggested that I’m not in the business of betting on this (isn’t that another site anyway?). As I said, no-one actually knows.
    I suspect you are a partisan. I am not.
    For what it’s worth, though, I am the person who made a five figure sum betting that John Major’s Tories would win the 1992 general election, and that was nothing whatsoever to do with my subsequent meetings with him in the Cabinet Room and elsewhere in No 10 as an adviser on polling research.

    Perhaps you may have mistaken me for someone else, or something.

  19. Is it any longer possible to draw conclusions from previous trends in Labour/Conservative support levels?

    I find the volatility of the polls over the last year to indicate that the electorate is far more sensitive than before to image, press coverage, and politics in general.

    Ideological differences no longer guarantee a “core vote”, since they’ve largely disappeared, and policies are formed more and more according to what the policy makers judge the voters want, rather than trying to sell us a coherent set of ideas, and such judgements seem to be coming from analysis of the polls.

    In a similar vein, it is natural to be wary of making predictions – especially of events, and so you’ll quite often see closing remarks like “interesting times ahead!”

  20. Robert Waller

    Don’t spoil my fun…you could have pretended to be the Robert Waller I am thinking of for just a little while!…he is probably playing his harp somewhere I guess….you hardly upset me or offended me with your opinions Robert but you cannot virtually dismiss the Conservatives better than sporting chance of winning the next election outright without expecting somebody to tweak your tail. I don’t know if your prognosis is based on the ‘uniform swing’ argument ie that the Tories have to be 10 points ahead in the polls before scenting a majority. This view has been debated many times on this site and by less allegedly ‘partisan’ contributors than I with the consensus being that a uniform swing is most unlikely, indeed I contend that an unravelling of the anti Tory tactical vote is well in progress. I would say that the last 4 general elections offer less of a guide as to what might happen if The Tories were to win by that margin than the 1987 election. Above all else if the Tories are heading for victory it is not because they are so popular-they are not- but because the government is so widely despised even by their own supporters. People won’t vote for a Mr Bean.

  21. If 32% say they will now, and the coverage can’t get much worse, then it’s likely that Mr could end up just as popular as Mr Snake or Mr Giraffe.

  22. Nick
    First of all, I am pretty certain that there isn’t another Robert Waller who said anything about the 1970 election – it must have been someone with a different name. It is true that a lot of people called that one wrong – including ‘yer darling Harold’ himself, of course! – but even if I had been old enough I doubt I’d have been among them as I did become interested in elections during the great byelection blitz of 1968-69.
    In a way it’s partly that experience that guides my tentative – and I repeat, if you look at the way I presented my non-predictions originally, you cannot in fairness call this ‘dismissing the C chances’ of an overall majority. Historically I think Oppositions have been much further ahead in the mid-term slump period; sometimes they have won, sometimes lost.
    On uniform swing, most elections have conformed fairly closely to this pattern overall and it tends to be the wishful thinkers who ‘dismiss’ it. Parties may do slightly better in marginals (usually incumbents) but there is unlikely to be anything like the end of the way the electoral system is currently biased against the Conservatives.
    That is because the main reasons why the C need to be around 8 or 9% ahead for an overall majority will still pertain:
    1. The over-representation of Wales and to some extent inner city seats.
    2. The lower turnout in Labour seats – a very striking feature.
    If the turnout rises, this may benefit Labour as it was largely their supporters who did not vote in 2001 and 2005, whne there was no real chance of them losing.
    It’s not really so much the boundaries or alleged tactical voting.
    In sum, the Conservatives could win an overall majority, but there is really rather little evidence that it will happen at present, just partisanship; it’s a mountain to climb. I know few impartial observers or level-headed Conservatives who think there is a high chance of it; most talk of victory in two leaps.
    There is an ancient philosophical problem, is there not, about distinguishing the ‘is’ and the ‘ought’.

  23. Robert Waller
    Oh yes indeed there was a pundit bearing your name way back in time but never mind I am more interested in your political points to which I respond as follows
    1 You make no comment on my point about the government’s popularity which with the current economic outlook hardly looks set to improve. Oppositions don’t win elections governments lose them. That’s what happened in 1997 and that’s what looks to me increasingly likely to happenen whenever Brown takes the plunge in the next two years.As for turnout I completely disagree that the liklehood of defeat increases the chances of Labour supporters turning out. Opposition supporters especially Tories are much more motivated when a government becomes loathed as this one is whereas government supporters in this case Labour become more not less lethargic with time with the memories of how they once hated the Tories having faded to the point of indifference.
    2 The longer a PM in this situation clings to office the greater the feeling of time for a change will grip the country as in 1945,1951,1979 and 1997.
    3 The Lib Dems are weaker than at any time in the last 20 years. one third of their vote has deserted back to the Tories. It is their weakness plus that of UKIP which will help the Tories to register the swings they need in many of the seats they need to seize
    4 The number of Welsh seats is comparitively small and the Tories have never even in their halcyon days commanded a majority of them .It is not an impediment to Tory victory.Your point about inner city seats is true but to some extent it has ever been thus.
    5 In my constituency lots of middle class voters have not bothered to vote at all in years because they felt that the Tories were unelectable and that Labour if not exactly their cup of tea were running the economy in a way that did not affect their well being. I can tell you that all that began to change over a year ago. They will vote en masse next time Tory to get Labour out because they are angry. Very angry.

  24. Nick , whilst I agree it is unlikely that Brown and Labour will recover from their current levels of unpopularity , it is not impossible . The Conservatives were even more unpopular in 1981 , at times in 1985/86 and in 1990 and won the subsequent general elections .
    Your statement that LibDems have lost 1/3rd of their vote to the Conservatives is plainly false . The latest polls show ( comparing vote at the last GE with intention now )
    Populus 1/7th
    ICM 1/5th
    Comres 1/6th
    with in all cases a small compensatory movement from those who voted Conservative in 2005 but say they will vote LibDem now .
    I find no evidence talking to people at work or in the pub that voters en masse are angry with Labour . Certainly they are unhappy with things at the moment and have been for some time but angry – no . The only issue that excited any real interest and political discussion for most of them was the lost CD data . Other than that most people are continuing to live in a state of political apathy not too happy with Labour but not enthusiastic over any other party either .

  25. Nick,

    Robert is quite right – there are some parts of the electoral “bias” towards Labour that are pretty much set in stone, like Welsh over-representation and for all intents and purposes the very low turnout in inner city slum seats.

    Increased turnout might well help Labour as Robert says, but it’s possible it might actually reduce the appearance of bias in the system (imagine Gordon Brown enthused traditional Labour voters in safe inner city seats who hadn’t bothered voting in recent elections because the Conservatives were out of it and they didn’t like Blair. Labour will pile up extra votes for no return in seats, resulting in the electoral system appearing to become fairer).

    Other bits of the electoral ‘bias’ are more variable and dependent on public opinion, most importantly tactical voting. Martin Baxter demonstrates its importance using two wonderful graphs on his website here – they are of seats where the Conservatives have between 35% and 40% of the vote, and seats where Labour have between 35% and 40% of the vote. Labour hold lots of theirs because the opposition vote is split. The Tories hold few of theirs because voters have unifed behind the party best able to beat them. Obviously, should the Tories become more popular and Labour less so this can change… but it won’t change things that much.

    On a straight uniform swing the Tories need to be 11% ahead. In reality my guess is they could do it on a smaller lead than that, because of disproportionate swings and changes in the pattern of tactical voting, say 7% or 8% or so… but Robert just said that he thought they could do it on 8% or 9% so it’s not like there’s a huge difference in opinion. If the Tories get just a small lead over Labour then, unless the Lib Dems have been completely smashed, they are NOT going to have a majority, and probably won’t be the largest party.

  26. Mark – I think 1985/6 is the more useful example. A 1981 style recovery would require a popular war and the main opposition splitting into rival parties, a 1990 style recovery would need a change in leadership. 85/86 is a better example of a government recovering without some major event to provoke it (though of course, who’s to say there won’t be an ‘event’ that saves Labour?)

  27. This doesn’t look like Dr Robert Waller’s style actually, but I apologise if it is.

  28. Actually I think it is,

    Robert I think I disagree with you on the likelihood
    of defeat not urging Labour supporters to turn out.
    It wouldn’t be on the scale of Tories appearing from under stones in April 1992,
    but surely there are at least some Labour non voters of 2001/2005 and basically Labour but Lib Dem 2005 voters who would rally to the polls if the Tories looked like winning or doing well.

  29. Sorry I am disagreeing with Nick Keene, not Robert Waller.
    I’ll read all this properly later!

  30. Nick, don’t you think that I’d be in a better position than you to know if there are two people writing on this subject with my name (yes, Joe, it is me – I’ll say just two words – Twickenham Riverside!)? Still, I won’t be discourteous enough to assume that just because you’re wrong there that you’re wrong on everything else!
    so …
    1. How can you or anyone else be sure of what the economy will be doing in 2009/10? I am known for being accused of economic determinism myself, but the best evidence I have (from quite an authoritative source, but still an economist!) is that our best guess is a slowdown, but not a recession – which is consonant with a decline in Labour’s position since 2005, though not a wipe-out (the Tories will need about 90-100 gains from Labour for an overall majority).
    As for turnout, as I’ve said on an American thread here, it tends to increase as elections appear close. If there had been compulsory voting in 2001, God forbid, Labour might have been 15% plus ahead. You can tell this from polling and British Election Study analysis.
    Incidentally, that links with one of Anthony’s points: yes, a higher turnout might reduce the pro-Labour bias (while not helping the C), but I’d guess the differential between Lab and C seats would remain.
    On the economy in general as a guide, if you think it’s important, as I do, I have always suggested looking at the ‘best party on the economy’ question rather than voting intention, which I actually think isn’t very helpful (as a pollster myself from 1986 to 1998). The last I saw had the Conservatives 2 or 3% ahead, which I think is a pretty good stab at what would happen in a general election now. But the future? Unknowable!
    2. I could quibble with the selective figures you give eg Churchill in for only 5 years, Attlee 6, Callaghan 3, Major 7, Brown 2 or 3 … compared with the re-election of Thatcher after 8 in 1987 (an election you mentioned somewhere), Blair after 4 and 8 in 2001 and 2005 … but actually I agree time for a change is an independent variable which probably reduces Labour’s chances of a fourth term. It doesn’t get the Tories to an overall majority necessarily, though.
    3. whether the Lib Dems are weak now is arguable and irrelevant. How strong will they be in 2009/10? It just takes one of their byelection specials to get them back into the 20s in the polls. And such is their incumbency power that I myself think that even with say 13 or 14% of the vote they would not lose more than 10 or 15 seats. Now that IS one area where unoform swing does not apply!
    4. Ever thus and still will be. I’m not making the ludicrous point that Tories need to win inner city seats to win an overll majority – they don’t, no more than Labour needed seats in Surrey etc – but just to support the explanation of the electoral bias, and how it is likely to continue (which you doubted).
    5. Subjective, subjective. Turnout among the middle class has demonstrably been higher than working class in recent elections, probably more so than ever before in 2001-5. Angry? You may be. So may some others. But I try to quantify, to be dispassionate and as I’ve said, not to mix up what I may want and what I think is most likely to happen.
    Though I’d agree on one or two things: what does Robert waller know? Either of them?!

  31. I think the Lib Dems might have a tip over point.
    If they had just 13%, that would surely be below the point where lots of them would lose on a national swing (down over 9 points nationally).

    I doubt they will go as low as 13% though – 17/18% seems a more credible figure. I’d like them to be destroyed – but that’s another matter.

  32. I have done an analysis since the last election, based on a 6-poll rolling average and uniform probabilistic swings using Anthony’d notionals. An SNP 15% rise is assumed, together with a calculation of LibDem incumbency based on previous elections where their national vote declined.

    For a brief 1 week period in December 2007 the Tories crested to a forecast 15 seat majority, but now appear to have declined again. This was the first occasion since 1992 that a Tory majority of any kind was forecasted..
    Con/Lab left hand scale (blue/red lines); LibDem right-hand scale (yellow columns)

    Some interesting stats… Since May 2005 the “average” forecast is
    Con 254
    Lab 310
    LD 48
    Nats 17 (based on +15% poll assumption)
    Oth 3
    NI 13 (SF abstain)

    62% of the time a Hung Parliament has been forecast…
    36% of the time a Labour majority….
    2% of the time a Tory majority…….

    73% of the time Labour was forecast to be ahead of the Tories in seats….

    The Tories don’t appear to be doing well enough to expect even largest party status at the next election, never mind a majority. A hung-parliament remains odds-on, but we are now past the half-way stage of this parliament, and might expect the balance to start moving back to Labour…

  33. Mark Senior
    What do you mean by the latest polls? The latest polls were YouGov/ MORI/ICM not Populus or ComRES.
    Robert Waller
    Where did I suggest that there would be a recession? Nobody knows that one . What I suppose I am suggesting is that a number of factors are combining to reduce if not end the feel good fsctor literally as we speak.At the very least we have a slowdown possibly a slump in prospect and I won’t even dwell on the likes of inflation , public spending, house prices, consumer confidence , public and personal debt. A lot of chickens are coming home to roost and I think-unfortunately for all of us- that it will take longer than 2 years for all of them to get their necks wrung.
    I am not denying that there is an electoral bias against the Tories but there is nothing new about that.Nor are the predictions that the Tories can’t win as a result-I have heard it all before. Obviously if the opposition in a Labour held seat say with a 5,000 majority like Dover-no 90 on the Tory target list- continues not only to split their votes between a Tory challenger (16700), a third placed Lib Dem(7600) and a fringe party of the right ie UKIP(1,200) and even worse from the Tory point of view with others still voting tactically to keep the Tories out by supporting Labour then the Tories cannot prevail.However it would not surprise me in the least to see the Tories take this seat with a 1,500 majority or more having taken 5.000 votes or so right across the board from all of the above with votes moving in both directions between Labour and Lib Dem.I won’t dwell on turnout for who can predict that? I simply do not accept that it is wishful thinking to suggest that such an outcome cannot be repeated all over the country or that the Tories only have a 15% chance of getting a majority. It’s much better than that although I do agree with Robert on one thing -the Tories still have a considerable part of the mountain to climb but then of course I am sure that somewhere along the line the late Sir E Hillary and the Hon D Cameron must be related…

  34. Nick , neither Yougov nor Mori show a comparison with how people voted in 2005 although Mori do in some polls . The other 3 pollsters do in every poll and clearly show in all their recent polls nothing like 1/3rd of LibDem voters moving to the Conservatives and an admittedly small number moving in the opposite direction .

  35. I suspect the economy will avoid a recession (I hope it does) but it will be slow growth, and there will be concerns, and sectors in difficulty.

  36. To add a footnote to my last message and with no idea of the latest odds offered by the bookies I would at this stage give the Tories a 30% chance of a majority and (overall) a 60% chance of being the largest party. Those figures may or may not shorten as we get towards 2009 -but that’s for the future.

  37. Nick, apart from wishful thinking, what are you basing your probabilities on?

  38. Nick-these odds from BestBetting now:-

    Date of GE
    2008 Best14/1 – Worst 6/1
    2009 Best 1/1 – Worst 0.64/1
    2010 Best 1.38/1 – Worst 1.1/1

    Most Seats
    Con Best 0.84/1 – Worst 0.67/1
    Lab Best 1.22/1 – Worst 1.00/1

    Overall Maj

    None 1.75/1
    Con 1-25 8/1
    Lab 1-25 10/1

  39. RodCrosby


  40. The thing I always say in responce to the ‘opposition has to be miles ahead int he mid-term’ argument is that that only applies when the Conservatives are incumbant and Labour in Opposition. There are examples of Labour being streets ahead in oppositions nd still losing- but never the other way around. The Tories in all of the last two elections did better int he election than in the mid term. There has never been a swing back to any incumbant Labour Government in an election which has overcome a Tory lead which was registered in the mid-term. I therefore don’t see how this point holds.

    One further thing. Given the still hopeless Tory scores in Scotland, they must be doing proportionatly even better in the rest of the UK by a few %. That might have some impact.

1 2