Labour Target Seats

These are the hundred and twenty five seats with the lowest percentage majorities over the Labour party. This does not necessarily mean they would be the most winnable seats for the Labour in practice, or that they are the seats the Labour party will actually be targetting at the next general election. The Labour party won 232 seats at the last general election, so would need to win another ninety-four seats to secure an overall majority in the House of Commons.

1. Derby North Majority 41 (0.1%)
2. Gower Majority 27 (0.1%)
3. Croydon Central Majority 165 (0.3%)
4. Vale of Clwyd Majority 237 (0.6%)
5. Bury North Majority 378 (0.8%)
6. Morley & Outwood Majority 422 (0.9%)
7. Plymouth Sutton & Devonport Majority 523 (1.1%)
8. Thurrock Majority 536 (1.1%)
9. Brighton, Kemptown Majority 690 (1.5%)
10. Bolton West Majority 801 (1.6%)
11. Telford Majority 730 (1.8%)
12. Weaver Vale Majority 806 (1.8%)
13. Bedford Majority 1097 (2.4%)
14. Plymouth Moor View Majority 1026 (2.4%)
15. Lincoln Majority 1443 (3%)
16. Cardiff North Majority 2137 (4.1%)
17. Peterborough Majority 1925 (4.1%)
18. Sheffield, Hallam Majority 2353 (4.2%)
19. Corby Majority 2412 (4.3%)
20. Waveney Majority 2408 (4.6%)
21. Warrington South Majority 2750 (4.6%)
22. Southampton, Itchen Majority 2316 (5.2%)
23. Keighley Majority 3053 (6.2%)
24. North Warwickshire Majority 2973 (6.3%)
25. Carlisle Majority 2774 (6.5%)
26. East Renfrewshire Majority 3718 (6.6%)
27. Leeds North West Majority 2907 (6.7%)
28. Halesowen & Rowley Regis Majority 3082 (7%)
29. Crewe & Nantwich Majority 3620 (7.3%)
30. Erewash Majority 3584 (7.4%)
31. Hendon Majority 3724 (7.5%)
32. Ipswich Majority 3733 (7.7%)
33. Broxtowe Majority 4287 (8%)
34. Stroud Majority 4866 (8%)
35. Calder Valley Majority 4427 (8.2%)
36. Northampton North Majority 3245 (8.3%)
37. Blackpool North & Cleveleys Majority 3340 (8.4%)
38. Pudsey Majority 4501 (8.8%)
39. Sherwood Majority 4647 (9.1%)
40. Amber Valley Majority 4205 (9.2%)
41. Colne Valley Majority 5378 (9.4%)
42. Hastings & Rye Majority 4796 (9.4%)
43. Bristol North West Majority 4944 (9.5%)
44. Edinburgh North & Leith Majority 5597 (9.6%)
45. Harrow East Majority 4757 (9.7%)
46. High Peak Majority 4894 (9.7%)
47. Stockton South Majority 5046 (9.8%)
48. Northampton South Majority 3793 (9.8%)
49. Norwich North Majority 4463 (10.2%)
50. Stevenage Majority 4955 (10.3%)
51. Enfield, Southgate Majority 4753 (10.4%)
52. Cannock Chase Majority 4923 (10.5%)
53. Morecambe & Lunesdale Majority 4590 (10.6%)
54. Nuneaton Majority 4882 (10.6%)
55. Dudley South Majority 4270 (11.2%)
56. Finchley & Golders Green Majority 5662 (11.2%)
57. South Ribble Majority 5945 (11.3%)
58. Worcester Majority 5646 (11.3%)
59. Rossendale & Darwen Majority 5654 (11.5%)
60. East Lothian Majority 6803 (11.5%)
61. South Swindon Majority 5785 (11.7%)
62. Southport Majority 1322 (3%)*
63. Preseli Pembrokeshire Majority 4969 (12.3%)
64. Paisley & Renfrewshire South Majority 5684 (12.3%)
65. Pendle Majority 5453 (12.3%)
66. Dover Majority 6294 (12.6%)
67. Reading East Majority 6520 (12.9%)
68. Warwick & Leamington Majority 6606 (13%)
69. Scarborough & Whitby Majority 6200 (13%)
70. Aberconwy Majority 3999 (13.3%)
71. Crawley Majority 6526 (13.4%)
72. Vale of Glamorgan Majority 6880 (13.4%)
73. Arfon Majority 3668 (13.6%)
74. Gloucester Majority 7251 (13.7%)
75. Great Yarmouth Majority 6154 (13.8%)
76. Reading West Majority 6650 (13.8%)
77. Carmarthen East & Dinefwr Majority 5599 (14.2%)
78. South Thanet Majority 2812 (5.7%)*
79. Brighton, Pavilion Majority 7967 (14.5%)
80. Chipping Barnet Majority 7656 (14.5%)
81. Stourbridge Majority 6694 (14.5%)
82. Elmet & Rothwell Majority 8490 (14.7%)
83. Milton Keynes South Majority 8672 (14.7%)
84. Aberdeen South Majority 7230 (14.8%)
85. Carmarthen West & South Pembrokeshire Majority 6054 (15%)
86. Camborne & Redruth Majority 7004 (15.2%)
87. Portsmouth South Majority 5241 (12.5%)*
88. Battersea Majority 7938 (15.6%)
89. Edinburgh South West Majority 8135 (15.8%)
90. Redditch Majority 7054 (16%)
91. Gravesham Majority 8370 (16.7%)
92. Dumfries & Galloway Majority 6514 (11.5%)*
93. Milton Keynes North Majority 9753 (16.9%)
94. Rutherglen & Hamilton West Majority 9975 (17.4%)
95. Cleethorpes Majority 7893 (17.5%)
96. Watford Majority 9794 (17.5%)
97. Loughborough Majority 9183 (17.6%)
98. Ochil & South Perthshire Majority 10168 (17.6%)
99. Clwyd West Majority 6730 (17.7%)
100. Shrewsbury & Atcham Majority 9565 (17.7%)
101. Paisley & Renfrewshire North Majority 9076 (18%)
102. South Basildon & East Thurrock Majority 7691 (16.9%)*
103. Lanark & Hamilton East Majority 10100 (18.3%)
104. Canterbury Majority 9798 (18.4%)
105. Dunfermline & West Fife Majority 10352 (18.6%)
106. Kingswood Majority 9006 (18.7%)
107. Stafford Majority 9177 (18.8%)
108. Kirkcaldy & Cowdenbeath Majority 9974 (18.8%)
109. Harlow Majority 8350 (18.9%)
110. Shipley Majority 9624 (19%)
111. Chingford & Woodford Green Majority 8386 (19.1%)
112. Edinburgh East Majority 9106 (19.3%)
113. Glasgow Central Majority 7662 (19.4%)
114. Airdrie & Shotts Majority 8779 (19.8%)
115. Carshalton & Wallington Majority 1510 (3.2%)*
116. Filton & Bradley Stoke Majority 9838 (20.1%)
117. Stirling Majority 10480 (20.1%)
118. Midlothian Majority 9859 (20.4%)
119. Basingstoke Majority 11063 (20.9%)
120. Linlithgow & Falkirk East Majority 12934 (21%)
121. Bexleyheath & Crayford Majority 9192 (21.1%)
122. Kensington Majority 7361 (21.2%)
123. Rugby Majority 10345 (21.2%)
124. Ayr, Carrick & Cumnock Majority 11265 (21.5%)
125. Rochford & Southend East Majority 9476 (21.7%)
Comments - 3,588 Responses on “Labour Target seats”
  1. Okay, listen up.

    I’ve decided you will all benefit from some more of my unparalleled psephological acumen.

    Here goes…

    10 MONTHS UNTIL MAY 2015 GENERAL ELECTION:
    HISTORICAL PRECEDENT POINTS TO 7% TORY LEAD.

    As of today (July 7 2014) this parliament has exactly ten months to run.

    The Tories are VERY LIKELY to recover to a popular vote lead in the general election scheduled for May 7 next year. They are also LIKELY (though not VERY LIKELY) to win a plurality of seats. However, the balance of probability is that they are UNLIKELY to secure an overall majority.

    This is what historical polling tells us about the likely shift in voting intentions during the last ten months of this parliament.

    There have been eight previous post-war Conservative parliaments. On average the Tories held a lead of -6.76% exactly ten months before seeking to renew their mandate, but on general election day they went on to average a lead of +3.28%. This represents a pro-Tory ‘swing back’ of +5.02%.

    The latest poll of polls shown on the front page gives the Conservatives a lead of -3%. On this basis the ‘swing back’ implied by historical precedent would be enough to give them a +7% lead in the May 2015 General Election, on a par with May 2010. This would once again ensure them a plurality (though not an overall majority) of parliamentary seats.

    It is, however, necessary to sound a note of caution about this estimate because there was a lot of variation in the extent of the Tory recovery from one parliament to another: the ‘swing back’ was virtually negligible during the last ten months of the 1955-59 and 1979-83 parliaments, modest in the case of the 1951-55, 1959-64 and 1992-97 parliaments and substantial in the case of the 1970-74, 1986-87 and 1987-92 parliaments.

    There was a positive ‘swing back’ in seven of the eight parliaments. The only exception was 1955-59, where there was an infinitesimally tiny swing against the incumbent Tory government.

    In six of the eight parliaments (75%) the ‘swing back’ to the Tories was of a magnitude that would overwhelm Labour’s current popular vote lead, though in only five cases (62.5%) would the ‘swing back’ be sufficient to give the Tories a lead in terms of parliamentary seats (and in just three cases – 37.5% – enough to give them an overall majority).

    Here are the full tables…

    1951-55:
    Con lead as at 26.7.54: -6%
    Con lead at general election held on 26.5.55: +1.8%
    Con ‘swing back’: +3.9%

    1955-59:
    Con lead as at 8.12.58: +4.5%
    Con lead at general election held on 8.10.59: +4.2%
    Con ‘swing back’: -0.15%

    1959-64:
    Con lead as at 15.12.63: -10.85%
    Con lead at general election held on 15.10.64: -1.9%
    Con ‘swing back’: +4.475%

    1970-74:
    Con lead as at 28.04.73: -18.2%
    Con lead at general election held on 28.02.74: +0.6%
    Con ‘swing back’: +9.4%

    1979-83:
    Con lead as at 09.08.82: +12.5%
    Con lead at general election held on 09.06.83: +15.2%
    Con ‘swing back’: +1.35%

    1983-87:
    Con lead as at 11.08.86: -6%
    Con lead at general election held on 11.06.87: +11.7%
    Con ‘swing back’: +8.85%

    1987-92:
    Con lead as at 09.06.92: -10%
    Con lead at general election held on 09.04.92: +7.6%
    Con ‘swing back’: +8.8%

    1992-97:
    Con lead as at 01.07.96: -20%
    Con lead at general election held on 01.05.97: -13%
    Con ‘swing back’: +3.5%

    AVERAGE OF ALL EIGHT POST-WAR TORY PARLIAMENTS:
    Con lead ten months before end of parliament: -6.76%
    Con lead at subsequent general election: +3.28%
    Con ‘swing back’: +5.02%

    [NB: in each case the poll used is the one whose fieldwork coincides most closely with the precise date that fell ten months before the general election in question. If the mid-point of the fieldwork of two or more polls are equally close to that date then the polls are averaged. The figures are taken from Mark Pack’s data as previously posted on this website by Anthony Wells].

    Below I answer some of the criticisms which Barnaby and others have levelled at this ‘swing back’ theory:

    CRITICISM #1: With less than one year to go, there simply isn’t enough time for the Tories to turn it around.

    MY ANSWER: Simply not true, as the above tables prove.

    CRITICISM #2: ‘Swing back’ will not work this time because people who voted Tory in 2010 have not switched directly to Labour in significant numbers. This means that even if those defectors return home it will not be enough to propel the Tories to victory.

    MY ANSWER: The latest YouGov computer tables show that 7% of 2010 Tory voters have switched directly to Labour and a similar (6%) proportion of 2010 Labour voters have switched directly to the Tories. The fact that the Tories are currently holding their own in Tory vs. Labour terms relative to 2010 implies an inherent weakness in Labour’s position. It is important to understand that when an elector responds to a hypothetical voting intention question they can sometimes behave differently to the way they would when confronted with the reality of casting their vote in an actual general election. One of the most unexpected snap elections was the one which was suddenly called by Edward Heath in February 1974: as soon as it was announced an almost consistent Labour lead was replaced by an equally consistent Tory lead. This tells us that whatever the relative strengths of the two main parties now, the actuality of the 2015 General Election is likely to produce an automatic shift in favour of the government (though the fact that it will not be a ‘snap’ election may mean the shift is more gradual and may have largely taken place before the official starting gun is fired). That shift is likely to be brought about in part by a greater number of 2010 Labour voters switching directly to the Tories than vice versa. This trend of opposition voters switching directly to the governing party occurred on two of the three post-war occasions in which the Conservatives tried for a second term of office.

    CRITICISM #3: The Tories cannot hope to make net gains from Labour compared to 2010. After all, Gordon Brown’s recession-hit government was so unpopular.

    MY ANSWER: In terms of who would make the best prime minister David Cameron consistently held only single-digit leads over Gordon Brown during the weeks and months leading up to the 2010 General Election. In short, leader preference conformed closely to party affiliation. This is not the case today: Ed Miliband lags a full 14% behind Cameron in terms of perceived prime ministerial qualities according to the latest MORI poll. More ominously there is now a substantial minority of Labour voters who rate the Tory leader higher on this criteria. Furthermore the Conservatives now have a clear and consistent lead on the economy (by 33% to 21% according to the latest Opinium survey). In fact less than two-thirds of Labour supporters back their own party on this issue whereas nine-tenths of Tory voters think the Conservative team is better at handling the country’s finances. The coalition may have wasted a good few years by sticking to plan ‘A’ rather than opting for Labour’s plan ‘B’ but that fact has now been largely forgotten as growth slowly returns.

    CRITICISM #4: The ‘swing back’ theory will not work this time because Labour has taken a large slice of the Lib Dem vote and is likely to hold most of this due to Nick Clegg’s participation in the Tory-led coalition.

    MY ANSWER: Yes, that will be a modest drag on the Tory lead over Labour in May 2015 and will to a certain extent ‘cushion’ us from the effects of the growing lead which the Conservatives are enjoying on the key issues of leadership and perceived economic competence. However, it needs to be understood that this is already factored into the equation: the latest YouGov tables show that Lib Dem defectors are contributing 7 percentage points to Labour coffers and only 4 percentage points to Tory coffers. In other words they are the sole reason why we are still ahead. That net 3-point advantage for Labour will be at least partly maintained through to May 2015 and will make the election closer than it would otherwise be but it will probably be cancelled out by voters switching directly from Labour to Tory. It is worth remembering that Margaret Thatcher won an easy victory in 1987, despite the Harris exit poll showing clearly that most Alliance supporters would have voted Tory as their second choice in that year. In addition Tony Blair won in 2005, even though the Lib Dems drew their support disproportionately from Labour. From whom the centre parties take their votes can make a difference but it need not necessarily prevent an incumbent government from being re-elected.

    CRITICISM #5: It’s not true that all UKIP voters are ex-Tory, so any fall in UKIP support will not necessarily benefit the Conservatives.

    MY ANSWER: The latest YouGov tables suggest that UKIP has derived 6 percentage points of its support from people who voted Tory in 2010, versus little more than a single percentage point from 2010 Labour voters. They are clearly taking disproportionately from the Conservatives but in a general election scenario it seems possible that some of their support will return home, as happened in both 2005 and 2010. Should half of them go back to their 2010 allegiance then this alone would be enough to virtually demolish the current Labour lead.

    CRITICISM #6: Using historical precedent is inaccurate because in previous parliaments the polling methodology has been shown to be flawed.

    MY ANSWER: Simply not true. The only occasion in which the final poll was substantially out on the general election result was 1992. Admittedly in both that year and in 1997 it understated the Tory lead but in both February 1974 and 1983 it actually overstated that lead (albeit to a lesser extent). Across the eight post-war Conservative parliaments the Tory lead implied by the final poll averaged out at +2.99%, against an actual average Tory lead of +3.28% in the real vote. Even if we factor in this polling error the figures still imply a pro-Tory ‘swing back’ of +4.88%, which is only slightly less than the +5.02% shown in the above table. [NB: where two or more final polls had the joint latest fieldwork dates, they are averaged out].

    CRITICISM #7 (This is one I’m making myself so as to demonstrate the robustness of the data): The ‘swing back’ figures are distorted by the inclusion of the unusually large 18.2% Tory deficit recorded by NOP in their poll of 1-6 May 1973. This was easily the biggest lead enjoyed by Labour throughout that entire year.

    MY ANSWER: Yes, but that poll is the one whose fieldwork was closest to April 28th 1973 – which was exactly 10 months before the February 1974 General Election. To exclude it would be to interfere with the randomness of the sample of eight ‘swing backs’ and would make the resultant average a contrived figure. Nonetheless, for the sake of argument it’s worth calculating what the ‘swing back’ would be if we substituted that poll for the next nearest (just in case the data has been unduly affected by a statistical fluke): this happens to be the Gallup poll conducted from 10-13 May 1973, which gives a much smaller (5.5%) Tory deficit. Such a figure would imply a ‘swing back’ of only +3.05% in February 1974 (down from the +9.4% shown in our original simulation) and would have the effect of reducing the average ‘swing back’ for the eight parliaments from +5.02% to +4.23%. If we take the second nearest poll for ALL eight parliaments then the revised average ‘swing back’ becomes +4.3%. Though a little less than the +5.02% shown in the above table, this would still be enough to give the Tories a comfortable popular vote lead as well as a plurality in terms of parliamentary seats.

    CRITICISM #8: No two parliaments are the same and you can never tell what news events might be lurking round the corner.

    MY ANSWER: That’s absolutely right, which is why there is still everything to play for. Historical precedent can only give us a guide as to what is likely to happen because in psephology nothing is set in stone. Even so the trends shown by the underlying polling data are pretty clear and it seems inconceivable that the Labour lead can hold given how far behind we have fallen on the all-important criteria of leadership and perceived economic competence. Only the electoral system, combined with some as yet unforeseen news story, is likely to save us.

  2. Just one quick comment on the analysis. You appear not to have taken account of how in the past a Government was able to take advantage of a poll lead to choose its own polling day..
    Also re -70 – 74 I recall quite a few Tory leads in Autumn 73 – before the snap election was called.

  3. Agreed with Graham…in most of the other scenarios governments chose their own time to go to the country. I hope Robin Hood is around post May 2015, either to bask in the brilliance of his insights or to face the music in having been totally wrong.

  4. I’m sure Robin will be around. The main trouble with his analysis is it’s based purely on historical precedent; some of these are valid and should be duly taken into account, others fail to take into account the different situation we find ourselves in at present. Pessimism is fine if it’s based entirely on fact; when it’s conjecture dressed up as a scholarly thesis it’s not a lot of use in forecasting election results. There’s no substitute for gauging the mood as it is at present, and as it’s likely to be in a few months’ time, and for understanding the marginal seats which will decide the outcome. If you don’t either have or build up the knowledge of the seats where the election will be won or lost, you can’t make an accurate forecast.

  5. Indeed. There is also regional difference to be taken into account and the fact that we have never had the LD’s in government before, nor a fourth party where we simply cannot guess where the votes are going to come from.

    On top of that, the idea that there is always a swing back to government belies the fact that movement directly between the major two parties has been remarkably static. Even though policies sometimes seem triangulated, support does not appear to be any more fluid between them.

    I would suggest it is also because both main parties are retaining strength in areas strong for them already, but in particular the Tories in the south-east

  6. Hate to repeat myself yet again, but Robin appears to be predicting Con most seats but without a majority. His biggest error is not that prediction, which is quite plausible, it is his assumption that this will result in the Tories remaining in government. If the numbers stack up it is very possible that Labour could enter government even if they have fewer votes or even fewer seats than the Conservatives.

    I do detect a subtle shift within Robin’s faux scholarly thesis – not long ago he was telling us that the Tories were bound to “win” the election….now it is just Con most seats, short of a majority. Perhaps his predictions will gravitate further towards the consensus in the next 10 months to avoid embarrassment next May? 🙂

  7. This parliament has been very weird…very static.

    steady drift for the next 10 months….labour will drift down and tories will pick up VI leading to an inconclusive result…can’t see labour getting much fewer than 300 or the Tories more than 285…lib dems 32-37…something like that. This is pretty much the consensus view of betting markets and indeed of many posters on this site…BUT actually labour majority is looking like good value at 5-2….if anything bad happens to the economy in the next 10 months, the tories will be stuffed. the negatives for labour are priced in, I think….we’ve seen ed m for nearly 4 years… and none of the other main party leaders have changed in that time….Major’s greatest asset in ’92 was his novelty…. we forget how popular he was between Maggie’s downfall in Nov ’90 and the April ’92 election.

    Smithson, as always, is good on this.

    http://www1.politicalbetting.com/index.php/archives/2014/07/07/dave-is-beating-ed-on-leader-ratings-but-nothing-like-on-the-scale-of-major-over-kinnock-ahead-of-ge1992/

  8. You yourself HH have frequently pointed out that no main governing party in living memory has gone a full term in office and then increased its share of the vote, not even the fabled Margaret Thatcher or Tony Blair governments. It’s all very well to invoke historical precedent, but this is a very major one & seems pretty convincing, especially if UKIP poll so much more than in 2010 which everyone now accepts they will do. I can see the Tories polling a maximum of 37% for this reason which surely cannot allow for a lead over Labour of any more than 5% at its most optimistic from a Conservative point of view. No good making historical precedent your basis, then ignoring what is arguably the most important one of all.

  9. I don’t disagree with anything you’ve said there. As it happens I think your prediction of Tories on 35-36 and Labour a few points behind is very plausible. Depending on distribution of votes and gains from Lib Dems, that might just be enough to narrowly give the Tories most seats, but only very narrowly. But I can’t see Labour being behind by enough to prevent a Lib-Lab type coalition, as was the case in 2010.

  10. My current thinking, as I said above, is a Tory lead of between 2 & 3% over Labour. That would almost certainly mean Labour largest party.

  11. Ladbrokes’ latest betting suggests 11 Labour gains will come in the North West.
    http://ow.ly/yQTSZ

  12. I tend to look at seats first and VI later. I know this seems like a “backwards” way of doing it but it actually reflects what the election is. The absurdity of looking at national VI where constituencies or, in the case of the US, States are involved is shown by the US elections….VI can cover a multitude of sins.

    I am more confident of Labour reaching 295 seats than i am of them getting 32%, 33%, 34, 35 or 36% of the vote…a lot of the final VI depends on a) the UKIP vote and b) differential turnout in different areas.

    11 Labour gains in the NW does imply a labour gain of about 40-50 which feels about right.

  13. Again that’s a very good post.

    We’re very likely to see nothing short of a constitutional outrage at the next election – a party which finishes 2-3% behind the leader forming the government, perhaps even without needing to go into coalition.

    Together with the economic problems which will reappear after 2015, the savagery bestowed on the new government will make 1992-97 look like the teddy bear’s picnic. It will change politics forever and most likely finally drag the Tories away from supporting FPTP.

  14. It’s possible there’ll be a double whammy at the next election in terms of parties getting more votes but less seats. UKIP might receive more votes than the LDs and get zero seats compared to 35-40 for the LDs.

  15. In answer to your various comments, yes I will definitely be around in May 2015 – and having just watched the performance of Stewart Jackson (Con, Peterborough, current majority: 10%) on “The Daily Politics” I would be delighted if my forecast is proved wrong.

    In fact if I am wrong please all of you do please feel free to crack a farm’s worth of rotten eggs over my face – for to do so would in no way take away any of my delight at seeing us unseat some absolutely awful MPs.

    To deal with Graham’s specific point, however: yes, previous governments were able to go to the country at a time of their choosing and that certainly is one factor that makes the 2015 election different.

    However, I’m not sure that factor will be too important in determining the outcome next year and here’s why…

    Would Gordon Brown have held on by going to the country in the autumn of 2009? I doubt it.

    Would Michael Howard have won if Tony Blair had called the 2005 election in the autumn (rather than the spring) of that year? Maybe, but probably not.

    Would Tony Blair have lost had he delayed the 2001 election until 2002? Certainly not.

    Come to that would John Major have won by going to the country in 1996 instead of 1997? No way. Would he have won by going to the country in the autumn of 1991 instead of April 1992 – not sure it would have made much difference, but as his majority was quite slim in 1992 who knows?

    Then again, would Mrs Thatcher have lost by delaying the 1987 election until 1988 – nope. Or holding off the 1983 election until 1984..? Definitely not.

    In fact you have to go back to Jim Callaghan’s decision not to call an election in the autumn of 1978 to find an example of how the election date MIGHT have made the difference to the outcome, but even that does not prove Graham’s point: after all, Callaghan had the option of choosing the election date (just as Wilson did in 1970) and a fat lot of good it did him.

    Most post-war general elections were decided by well-worn trends relating to the public’s perception of the competence and general fitness to govern of the two main parties, rather than by the precise date on which the election was called.

    On a general note, would people like me to try and repeat this exercise 9, 8, 7 and 6 months etc., before May 2015? It’s your call.

  16. Oh, nearly forgot – Graham says he recalls quite a few Tory leads in the autumn of 1973. Sorry, Graham, but your memory has failed you.

    There were 25 published polls in 1973 and the only one which the Tories led on was the NOP poll conducted between 9-15 January of that year: it yielded a Conservative lead of +0.4%.

    Throughout the autumn of 1973 (Sept/Oct/Nov) there were nine polls ranging from Labour leads of 1.2% up to 9.5%. The Labour lead averaged 5.1% across that season.

  17. Tbh…i think macmillan would have had a lower majority in 1960, Eden a lower majority in 1956…i think macmillan might have squeaked it, if he had risked an election in 1963….I think Thatcher’s majority would have been a lot less in 1984 and 1988…

    you can be minute and analytical when you want, Robin Hood, and then vague and hazy when you don’t.

    The elections of ’55, ’59, ’83 and ’87 were decisive. So i am not suggesting that an election held a year later would have overturned an outcome. What i am saying is that the government would not have performed nearly as well, if the election had been delayed.

    Seeing that we are in the first coalition government since the war, a slight change could result in a different government in 2015, which would not have been the case in ’56, ’60, ’84 etc.

  18. Peter, you say Thatcher’s majority would have been a lot less in 1984 and 1988. What do you base that supposition on – the fact that her poll lead was less in 84 and 88 than it had been in 83 and 87? Surely that’s a function of the fact that there were NOT elections in 84 and 88, for if she had scheduled elections for those two years Michael Foot might still have been Labour leader in 84 (he did, after all, only resign because he lost the 83 election) and the fabricated temporary economic boom would have been engineered to occur in 88 rather than 87.

  19. re. 1984 I think the falklands bounce would have faded, not entirely, but more so…I also think the SDP would have faded, helping labour… one cannot of course prove what didn’t happen…

    re. 1988, I think the overheating of the economy would have hindered the tories…this is clearly all counterfactual…

    Thatcher clearly thought i was right, as she wouldn’t have held the elections in ’83 or ’87 is she thought she would have done as well in ’84 and ’88…she enjoyed power too much. I suggest she went early because she thought, rightly in my view, that those were the best years for her…she got a 144 in 1983, as everyone knows, and 102 in 1987. so she was right.

  20. To (roughly) quote from Alan Clark’s history of the Tories:

    “The Great Storm of 16 October 1987 and the Black Monday stock market crash the following week were, in retrospect, the key turning point of Mrs Thatcher’s government. After those events, absolutely nothing seemed to go right for her, setting in motion a gradual but relentless decline, culminating in the events of November 1990 which resulted in her final downfall.”

    The Tories would certainly have won a 1988 election but certainly not with a 1987 level of majority (but probably with a bigger majority than Major in 92). Labour first took a slim poll lead in 1989 IIRC.

  21. If the Tories had won an election in 1988 rather than 1987, I’m certain Labour would have won in 1992/93.

  22. As I have posted before, the long parliament model of 5 years doesn’t augur well for incumbents….only two governments since the last war were returned after a parliament of longer than 4 years. The 1945-1950 labour govt. managed a majority of 6, after having been elected with a majority of 146 in 1945. The 1987-1992 conservative government managed a majority of 21, after getting elected with a majority of 102 in 1987….

    1964, 1979, 1997 and 2010 all saw tired governments booted out. By contrast 1974 and 1970 saw governments turfed out after 4 years or sol….but 1955, 1959, 1983, 1987, 2001 and 2005 saw successful re-elections after a four year parliament. The 2010-15 government, I fear, will not emulate either Attlee’s socialists in 1950 or Major’s Tories in 1992.

  23. Empirically, just looking at the events, (i am claiming no predictive power for this), both governments re-elected after 5 years had 100+ majorities in the house of commons. British elections are about swings…both those government had a lot of fat to protect them against the electoral winter, so to speak.

    1992 still saw the tories lose 42 seats…even though it’s presented as a triumphant victory.

  24. This will probably change due to fixed term parliaments. Previously, governments staggered on for five years only if they were doing badly in the polls.

  25. Yes I agree. Peter is making too much of the fixed term aspect, though I largely agree with his forecast for next year.

  26. I think 5 years is a long time…i can’t think of any jurisdictions where they have fixed elections lasting as long as 5 years…i think the tendency of govts. to do badly when they are trapped in the headlights of a general election which people know will happen is understated….

    One of the boldest moves in recent history regarding terms of elections was when Schroder changed the constitution and brought on the election of 2005 in Germany which was expected in 2006…He looked decisive and almost overturned a 20% deficit in the polls. I think the electorate gives the governing party a slight benefit in calling the election. I think Heath would have been slaughtered if he’d waited till ’75…I also think the same of Wilson, though I am less sure about this, if he had waited till March ’71.

  27. The US Senate elects by thirds with six-year terms, much too long in my view.

  28. I concur that Peter’s thesis on five-year parliaments is misguided. The fact those governments on five-year terms were defeated was not by virtue of them being in an extra year – they were on five-year terms precisely because the polls were bad and were hoping an extra year would pull something out for them, which in most cases doesn’t happen, so the polls stay bad and the government gets defeated.

    Still, I’m on record in saying I think four-year terms are far better.

  29. For me, 36% is probably a glass ceiling for the Tories in this coming election.

    With the possible but somewhat unlikely exception of centre right Scottish nationalists, I just don’t see which segment of society that did not vote for the Tories in 2010 will do so to such an extent as to offset their losses to UKIP this time around, nor can I see turnout significantly falling, nor can I see a rise in turnout benefitting the Tories (on the presumption that of the three biggest parties, Labour’s vote would have taken the most convincing to turn out in 2010).

    That said, I think UKIP’s rise will narrow the extent to which Labour’s popular vote translates into a lot more seats than the Tories. I say that because I think UKIP will take more votes off the Tories than Labour (althought I think it will be more of a 3:2 ratio than the 2:1 or 3:1 that some people have it down as), but will cost the Tories very few seats.

  30. Van Fleet-

    do we think that the tories/coalition would have won an election this year if they hadn’t bound themselves by the fixed five year terms….would they have looked at the polls say in March and decided to wait an extra year?

  31. That’s irrelevant. Your thesis was that somehow the election defeats of 64, 79, 97, and 2010, were down to the fact they were five-year terms, which is complete rubbish if I may say so.

    For one thing, Brown would have fallen to a far bigger defeat had he run in 2009, and whilst I claim no insight on the 96 polling, certainly an election in 95 would have been worse than the one in 97, so stands to reason 96 would still have been worse than 97.

    78 would certainly have been better for Callaghan than 79, but it’s the events of 78-79 which did it, not because of the mere fact the government was five years long.

    Like I said, I agree it’d be better to have a four-year term, but your thesis that five-year terms in of themselves were cause for this defeats is barmy.

  32. Sometimes I wonder if H.H’s blunt style is starting to rub off on me.

  33. My thesis is a little more subtle than that….obviously it wasn’t the 5 years that led those governments to defeat…they went to 5 years because the polls in year 4 were bad and they hoped, against experience, that they could stand a better chance if they waited….in nearly all cases, the turnaround they hoped for didn’t materialize.

    Now in this situation, I ask, if the current government had enjoyed that flexibility would they have sought re-election in 2014…I expect not…obviously this can’t be proved, but i suspect they would have reasoned, like their doomed predecessors, “let’s wait a year”, “let’s wait to let the recovery take hold more firmly and see if something turns up”…of course it never did…

    I think labour would have done better in the summer of 1949, pre the sterling devaluation which took place in September 1949, than they did in February 1950…[I believe the King asked them to have the election at that slightly barmy date]…interestingly in 1950 labour won the popular vote by 6% over the tories but only had 315 seats against 282 for the blues…the british have been living with the anomalies of FTPT for centuries obviating the likelihood, in my opinion, of any popular outcry if the results next year throw up seeming anomalies.

  34. I don’t know about popular outcry from the British public, but I strongly suspect that if a Labour Government forms on a “shaky” mandate in terms of popular vote the media will be screaming about it at a very high pitch (and of course ignoring the fact that they supported FPTP and that this outcome would largely be down to the anti-Tory vote being better targeted and low turnout in Labour heartlands).

  35. It could be a very unstable government the next one, whoever does indeed form it.

    If the numbers have the Tories narrowly in front in terms of votes but not in seats, that could well allow Labour the opportunity to form a minority government if the Lib Dems refused to support the Tories again. I couldn’t imagine Ed Miliband wanting to go into coalition with the Lib Dems after everything that’s happened TBH, but who knows?

    There is every chance that if it is another Hung Parliament that we could be in for a very short Parliament and that an emergency election would have to take place within six months, a bit like the situation between February and October 1974. I think it is important that someone ends up with a majority, because after the slump in popularity that the Lib Dems have suffered over the past four years, if they end up in government again they’ll just become even more unpopular, as if they weren’t already. I think it’s for their own good as a party that they leave government and rebuild their image and voter base. So in those circumstances, short-term it would have to be either a minority Labour or Tory government I’m afraid.

  36. ‘I couldn’t imagine Ed Miliband wanting to go into coalition with the Lib Dems after everything that’s happened TBH, but who knows?’

    Oh he could. So can I. Indeed, so do most members of the party, grassroots up. We’re sensible enough to know that there’s no point ruling out a coalition if that’s what we have to do to kick the Conservatives out.

    What we would however want is no Clegg.

    ‘There is every chance that if it is another Hung Parliament that we could be in for a very short Parliament and that an emergency election would have to take place within six months’

    As we have to keep reminding people, we have fixed-terms now, with very little leeway for an early election. The 2011 law even proscribes against the use of the royal prerogative, even if everything else fails. Either two-thirds of the Commons votes for an early election, or a motion of no-confidence is passed.

  37. ”Oh he could. So can I. Indeed, so do most members of the party, grassroots up. We’re sensible enough to know that there’s no point ruling out a coalition if that’s what we have to do to kick the Conservatives out.”

    He wouldn’t and he shouldn’t. It’s the last thing EM would be best to do. Given all the slating he gets from the media as it is, can you imagine how much worse it would be for him if he entered government with Nick Clegg?

    ”As we have to keep reminding people, we have fixed-terms now, with very little leeway for an early election. The 2011 law even proscribes against the use of the royal prerogative, even if everything else fails. Either two-thirds of the Commons votes for an early election, or a motion of no-confidence is passed.”

    But what if a minority government was defeated on vote after vote, wouldn’t they hold their hands up and go back to the electorate, and just admit that they couldn’t carry on governing because there was no stable Parliament in prospect? We don’t want to go back to the days of confidence or no-confidence motions like there were as recently in 1992-1997 ever again.

  38. parliament could vote to repeal the fixed term act and probably would if the house of commons was in the situation you describe.

  39. “He wouldn’t and he shouldn’t. It’s the last thing EM would be best to do. Given all the slating he gets from the media as it is, can you imagine how much worse it would be for him if he entered government with Nick Clegg?”

    Hence why I did say we’d push for no Clegg. Not that we wouldn’t have good reason to, if only because he pushed for no Brown in 2010. To be honest though, I really would think Clegg would step aside without being pushed – even he would surely accept it’s perverse being a deputy to both a Conservative and Labour government.

    But like I said, if it’s a choice between a coalition with the Lib Dems, and the Tories staying in, don’t hesitate to decide which we’d take.

    “But what if a minority government was defeated on vote after vote, wouldn’t they hold their hands up and go back to the electorate, and just admit that they couldn’t carry on governing because there was no stable Parliament in prospect? We don’t want to go back to the days of confidence or no-confidence motions like there were as recently in 1992-1997 ever again.”

    I’m not saying it’s right, I’m just saying what the law says.

  40. “even he would surely accept it’s perverse being a deputy to both a Conservative and Labour government.”

    And as a supplementary point, I know that kind of thing does happen in PR countries. However, the United Kingdom has no culture of acceptance for such things. That’s what hundreds of years of FPTP does.

  41. ”Hence why I did say we’d push for no Clegg. Not that we wouldn’t have good reason to, if only because he pushed for no Brown in 2010. To be honest though, I really would think Clegg would step aside without being pushed – even he would surely accept it’s perverse being a deputy to both a Conservative and Labour government.

    But like I said, if it’s a choice between a coalition with the Lib Dems, and the Tories staying in, don’t hesitate to decide which we’d take.”

    But that’s the problem- Do you as a Labour supporter (I think you are) not mind at all to be associated with the Lib Dems, only as long as Nick Clegg is no longer the leader- I mean, surely it stands to reason, that after being in with the Tories, you wouldn’t surely want to go with a party that had betrayed the people that had voted for them, surely?

  42. More than a supporter, I’m a member, and an active campaigner. I’ve seen in action the dirty tricks the Lib Dems are often famed for. I have no love for alot of them.

    But, if there’s something I don’t like more than the Lib Dems, it’s the Conservatives with another term. If the arithmetic of a hung parliament demands so, the party will accept being in coalition with them.

  43. ”And as a supplementary point, I know that kind of thing does happen in PR countries. However, the United Kingdom has no culture of acceptance for such things. That’s what hundreds of years of FPTP does.”

    That’s why it be would still be strange if Labour went with the Lib Dems in any case. After allying themselves to the Tories in one Parliament, how can they just turn around and say, ”oh no it doesn’t matter anymore, we’ll go with Labour”, as if nothing ever happened- In politics, surely it stands to reason that the decisions you make are ones you have to stand by once you’ve made them. Call a spade a spade.

  44. To put my position very crudely (and probably a lot of Tories in the 2010-15 parliament) – if we’re going to face a firing squad, I’d rather the Lib Dems were standing in front of us.

  45. “That’s why it be would still be strange if Labour went with the Lib Dems in any case. After allying themselves to the Tories in one Parliament, how can they just turn around and say, ”oh no it doesn’t matter anymore, we’ll go with Labour”, as if nothing ever happened- In politics, surely it stands to reason that the decisions you make are ones you have to stand by once you’ve made them. Call a spade a spade.”

    Remember, this is the Lib Dems we’re talking about. They’ve wanted PR, and are quite happy to do politics as many of our continental brethren do, which often means doing a complete 180 and siding with your once-opponents.

  46. ”More than a supporter, I’m a member, and an active campaigner. I’ve seen in action the dirty tricks the Lib Dems are often famed for. I have no love for alot of them.
    But, if there’s something I don’t like more than the Lib Dems, it’s the Conservatives with another term. If the arithmetic of a hung parliament demands so, the party will accept being in coalition with them.”

    Oh, for goodness sake, you’ve basically just said you don’t like the Lib Dems. So how on earth could you stand to be in government with them?! And anyway, whose to say your party couldn’t go it alone? Who needs other parties? Sometimes in life, there comes a time when you have to stand on your own two feet, physically and metaphorically. The Labour Party is going to have to stand up, be counted and sail the ship alone if they’re going to prove anything to anyone, and that is what will do EM’s reputation the world of good if he does have a chance of power after next year’s election. (I don’t support Labour myself, incidentally).

  47. I think there’s a rather sly unspoken acknowledgement among many in Labour that the Lib Dems play nice and do as they’re told so would be pliable to Labour’s will in a coalition. However much more common is a desire for a majority. Coalition only if absolutely necessary to evict the Tories.

  48. ”Remember, this is the Lib Dems we’re talking about. They’ve wanted PR, and are quite happy to do politics as many of our continental brethren do, which often means doing a complete 180 and siding with your once-opponents.”

    Don’t entertain them then. If you’ve got a way about you, and are independent enough, you will ignore them, like the proverbial school bully. This is a kind of politics we can do without. I shudder to think what my forefathers would have made of the kind of political atmosphere we now find ourselves in…

  49. Well obviously I’d want a majority, or run as a minority government. To think I would want otherwise is silly.

    But if the arithmetic demands, we will have a coalition. I could stand to have them in government, because frankly the thought that we would give the Conservatives another five years to damage (my personal view, obviously) the country, just because we were acting like children and couldn’t stand to be in the same room as the Lib Dems, is far more unpalatable.

  50. ”I think there’s a rather sly unspoken acknowledgement among many in Labour that the Lib Dems play nice and do as they’re told so would be pliable to Labour’s will in a coalition. However much more common is a desire for a majority. Coalition only if absolutely necessary to evict the Tories.”

    Please, please, please, please, dear God, not ANOTHER coalition! The Labour Party can win an overall majority if they put their mind to it I’m sure, whether they actually will or not, is another matter- It could even be the Conservatives who do it, in fact, it’s all in the balance.

Leave a Reply

NB: Before commenting please make sure you are familiar with the Comments Policy. UKPollingReport is a site for non-partisan discussion of polls.

You are not currently logged into UKPollingReport. Registration is not compulsory, but is strongly encouraged. Either login here, or register here (commenters who have previously registered on the Constituency Guide section of the site *should* be able to use their existing login)