Here at Forecast UK we’ve been consistently predicting a large number of losses for the Liberal Democrats at the next General Election (if the public votes as it is currently indicating in the polls). How realistic are those predictions?
The forecast of Lib Dem losses is made on the basis of a resurgent Conservative Party that is consistently polling at a level 10% higher than its share of the vote in the 2005 election. At the same time the Liberal Democrats have been trailing in the polls with levels of support well under their share of the vote four years ago. On those figures the Conservatives would improve their lead over the Lib Dems by around 15%, giving a 7.5% swing to the Tories. This kind of figure would yield to the Conservatives almost 30 Lib Dem seats, as the table below indicates.
Lib Dem / Conservative Marginals 2009/10 Election
| Constituency | Electorate | Conservative | Lib. Dem. | Majority | Majority% |
|---|---|---|---|---|---|
| Romsey CC | 72177 | 22340 | 22465 | 125 | 0.25 |
| Solihull BC | 77910 | 20617 | 20896 | 279 | 0.53 |
| Westmorland and Lonsdale CC | 69363 | 22302 | 22569 | 267 | 0.54 |
| Taunton CC | 85466 | 25191 | 25764 | 573 | 0.96 |
| Eastleigh BC | 76844 | 18648 | 19216 | 568 | 1.14 |
| Somerton and Frome CC | 77806 | 22947 | 23759 | 812 | 1.50 |
| Hereford CC | 71813 | 19323 | 20285 | 962 | 2.05 |
| Carshalton and Wallington BC | 67844 | 16289 | 17357 | 1068 | 2.48 |
| Torbay BC | 76474 | 17288 | 19317 | 2029 | 4.29 |
| Cheltenham BC | 71541 | 15819 | 18122 | 2303 | 5.28 |
| Cornwall North CC | 86841 | 20766 | 23842 | 3076 | 5.49 |
| Sutton and Cheam BC | 63319 | 16922 | 19768 | 2846 | 6.79 |
| Richmond Park BC | 70555 | 20280 | 24011 | 3731 | 7.26 |
| Portsmouth South BC | 70969 | 13685 | 17047 | 3362 | 8.33 |
| Cheadle BC | 68123 | 19169 | 23189 | 4020 | 8.47 |
| Southport BC | 67977 | 15255 | 19093 | 3838 | 9.32 |
| Devon North CC | 76203 | 18868 | 23840 | 4972 | 9.57 |
| Bath BC | 66824 | 15463 | 20101 | 4638 | 10.12 |
| Brecon and Radnorshire CC | 55171 | 13277 | 17182 | 3905 | 10.18 |
| Teignbridge CC | 88674 | 21593 | 27808 | 6215 | 10.21 |
| Winchester CC | 85810 | 23749 | 31222 | 7473 | 12.12 |
| Dorset Mid and Poole North CC | 65924 | 16518 | 22000 | 5482 | 12.14 |
| Cornwall South East CC | 80704 | 18479 | 24986 | 6507 | 12.17 |
| Berwickshire, Roxburgh and Selkirk CC | 71702 | 13092 | 18993 | 5901 | 13.00 |
| Argyll and Bute CC | 67325 | 10150 | 15786 | 5636 | 13.04 |
| Colchester BC | 79010 | 14868 | 21145 | 6277 | 13.98 |
| Truro and St Austell CC | 80256 | 16686 | 24089 | 7403 | 14.36 |
| Oxford West and Abingdon CC | 80195 | 16653 | 24336 | 7683 | 14.61 |
| Yeovil CC | 77668 | 17096 | 25658 | 8562 | 17.15 |
| West Aberdeenshire and Kincardine CC | 65548 | 11814 | 19285 | 7471 | 17.94 |
| Norfolk North CC | 80784 | 20909 | 31515 | 10606 | 17.99 |
| Kingston and Surbiton BC | 72671 | 16431 | 25397 | 8966 | 18.02 |
| Lewes CC | 67073 | 15902 | 24376 | 8474 | 18.20 |
| Northavon CC | 81800 | 19839 | 30872 | 11033 | 18.68 |
| Twickenham BC | 72015 | 16731 | 26696 | 9965 | 19.28 |
| Hazel Grove CC | 64376 | 11607 | 19355 | 7748 | 19.81 |
| Sheffield Hallam CC | 59606 | 12028 | 20710 | 8682 | 21.42 |
| St Ives CC | 74716 | 13968 | 25577 | 11609 | 23.03 |
| Montgomeryshire CC | 46766 | 8246 | 15419 | 7173 | 23.83 |
| Berwick-upon-Tweed CC | 56944 | 10420 | 19052 | 8632 | 23.92 |
| Harrogate and Knaresborough CC | 65622 | 13684 | 24113 | 10429 | 24.33 |
| Edinburgh West BC | 65741 | 8817 | 22417 | 13600 | 30.05 |
| North East Fife CC | 62057 | 7517 | 20088 | 12571 | 32.60 |
Of course, this assumes a uniform swing across the country. In reality individual constituencies behave in a unique manner and the movement in votes is different depending on local circumstances. To analyse this, the table below shows the 43 seats where the Lib Dems won in 2001 and the Tories came second. Next to each seat is the 2005 result, showing who won, what the swing was and to who.
2005 Election – Swings in Con / Lib Dem Marginals
| Constituency | Electorate | Conservative | Lib. Dem. | Majority | Majority% | 2005 Win | Swing | Swing% |
|---|---|---|---|---|---|---|---|---|
| Newbury CC | 75903 | 26771 | 23311 | -3460 | 6.33 | Con | LDm to Con | 5.54 |
| Devon West and Torridge CC | 83489 | 25013 | 21777 | -3236 | 5.52 | Con | LDm to Con | 3.83 |
| Weston-Super-Mare CC | 74900 | 19804 | 17725 | -2079 | 4.23 | Con | LDm to Con | 2.48 |
| Ludlow CC | 64572 | 20979 | 18952 | -2027 | 4.36 | Con | LDm to Con | 4.07 |
| Guildford CC | 75566 | 22595 | 22248 | -347 | 0.67 | Con | LDm to Con | 0.9 |
| Romsey CC | 72177 | 22340 | 22465 | 125 | 0.25 | Lib Dem | LDm to Con | 2.32 |
| Eastleigh BC | 76844 | 18648 | 19216 | 568 | 1.14 | Lib Dem | LDm to Con | 2.64 |
| Somerton and Frome CC | 77806 | 22947 | 23759 | 812 | 1.5 | Lib Dem | Con to LDm | 0.12 |
| Hereford CC | 71813 | 19323 | 20285 | 962 | 2.05 | Lib Dem | LDm to Con | 0.06 |
| Carshalton and Wallington BC | 67844 | 16289 | 17357 | 1068 | 2.48 | Lib Dem | LDm to Con | 4.36 |
| Torbay BC | 76474 | 17288 | 19317 | 2029 | 4.29 | Lib Dem | LDm to Con | 4.91 |
| Cheltenham BC | 71541 | 15819 | 18122 | 2303 | 5.28 | Lib Dem | LDm to Con | 3.64 |
| Sutton and Cheam BC | 63319 | 16922 | 19768 | 2846 | 6.79 | Lib Dem | LDm to Con | 2.02 |
| Cornwall North CC | 86841 | 20766 | 23842 | 3076 | 5.49 | Lib Dem | LDm to Con | 6.36 |
| Portsmouth South BC | 70969 | 13685 | 17047 | 3362 | 8.33 | Lib Dem | LDm to Con | 3.61 |
| Richmond Park BC | 70555 | 20280 | 24011 | 3731 | 7.26 | Lib Dem | LDm to Con | 1.42 |
| Southport BC | 67977 | 15255 | 19093 | 3838 | 9.32 | Lib Dem | Con to LDm | 1 |
| Brecon and Radnorshire CC | 55171 | 13277 | 17182 | 3905 | 10.18 | Lib Dem | Con to LDm | 4.09 |
| Cheadle BC | 68123 | 19169 | 23189 | 4020 | 8.47 | Lib Dem | Con to LDm | 4.2 |
| Bath BC | 66824 | 15463 | 20101 | 4638 | 10.12 | Lib Dem | LDm to Con | 5.63 |
| Devon North CC | 76203 | 18868 | 23840 | 4972 | 9.57 | Lib Dem | Con to LDm | 1.76 |
| Dorset Mid and Poole North CC | 65924 | 16518 | 22000 | 5482 | 12.14 | Lib Dem | Con to LDm | 5.63 |
| Berwickshire, Roxburgh and Selkirk CC | 71702 | 13092 | 18993 | 5901 | 13 | Lib Dem | LDm to Con | 6.68 |
| Teignbridge CC | 88674 | 21593 | 27808 | 6215 | 10.21 | Lib Dem | Con to LDm | 2.56 |
| Colchester BC | 79010 | 14868 | 21145 | 6277 | 13.98 | Lib Dem | Con to LDm | 0.64 |
| Cornwall South East CC | 80704 | 18479 | 24986 | 6507 | 12.17 | Lib Dem | Con to LDm | 0.89 |
| Montgomeryshire CC | 46766 | 8246 | 15419 | 7173 | 23.83 | Lib Dem | Con to LDm | 1.16 |
| Truro and St Austell CC | 80256 | 16686 | 24089 | 7403 | 14.36 | Lib Dem | LDm to Con | 0.84 |
| West Aberdeenshire and Kincardine CC | 65548 | 11814 | 19285 | 7471 | 17.94 | Lib Dem | Con to LDm | 2.3 |
| Winchester CC | 85810 | 23749 | 31222 | 7473 | 12.12 | Lib Dem | LDm to Con | 2.08 |
| Oxford West and Abingdon CC | 80195 | 16653 | 24336 | 7683 | 14.61 | Lib Dem | LDm to Con | 1.6 |
| Hazel Grove CC | 64376 | 11607 | 19355 | 7748 | 19.81 | Lib Dem | LDm to Con | 1.06 |
| Lewes CC | 67073 | 15902 | 24376 | 8474 | 18.2 | Lib Dem | LDm to Con | 1.58 |
| Yeovil CC | 77668 | 17096 | 25658 | 8562 | 17.15 | Lib Dem | Con to LDm | 4.5 |
| Berwick-upon-Tweed CC | 56944 | 10420 | 19052 | 8632 | 23.92 | Lib Dem | Con to LDm | 0.31 |
| Sheffield Hallam CC | 59606 | 12028 | 20710 | 8682 | 21.42 | Lib Dem | LDm to Con | 1.51 |
| Kingston and Surbiton BC | 72671 | 16431 | 25397 | 8966 | 18.02 | Lib Dem | LDm to Con | 6.95 |
| Twickenham BC | 72015 | 16731 | 26696 | 9965 | 19.28 | Lib Dem | Con to LDm | 1.98 |
| Harrogate and Knaresborough CC | 65622 | 13684 | 24113 | 10429 | 24.33 | Lib Dem | Con to LDm | 1.68 |
| Norfolk North CC | 80784 | 20909 | 31515 | 10606 | 17.99 | Lib Dem | Con to LDm | 8.56 |
| Northavon CC | 81800 | 19839 | 30872 | 11033 | 18.68 | Lib Dem | Con to LDm | 0.48 |
| St Ives CC | 74716 | 13968 | 25577 | 11609 | 23.03 | Lib Dem | Con to LDm | 1.31 |
| North East Fife CC | 62057 | 7517 | 20088 | 12571 | 32.6 | Lib Dem | Con to LDm | 2.99 |
Overall, the average swing to the Liberal Democrats across these 43 seats was 0.7%. Across the nation as a whole though there was a Conservative to Lib Dem swing of 1.56%. Putting these two figures together, we can see that in the 2005 General Election, in Lib Dem / Tory marginals the Conservatives actually achieved a swing to themselves that beat the national swing by over 2%. This demonstrates that the idea that Liberal Democrat incumbency produces greater local popularity votes is disproved by the hard numbers. At the 2005 election, where the Conservatives were in second place to the Liberal Democrats they did better then the national picture.
Now, add this information to the current poll figures (our latest forecast has the Conservatives on 43.6% and the Liberal Democrats on 18.5% – a swing since the last election of 7.4% which would topple a majority in a seat of 14.8%). Begin by going down the first of our two tables – the current polling figures indicate that every seat down to Oxford West would go. Add onto that our swing figures from the 2005 General Election in Con / Lib Dem target seats showing the difference from the national picture in these key battlegrounds and the projected swing could actually be in the region of 10% where it counts, enough to overturn majorities close to 20%. That kind of figure would take all but 7 of the 43 Lib Dem / Tory targets at the next election.
Even if we assume that the voting figures are more like 40 / 22 (Tory / Lib Dem) at the next election, that still gives us a swing of 3.8%. That would topple seats down to Richmond Park, but with including our 2005 battleground analysis we would also see seats like Teinbridge and Winchester going blue, half way down the list.
No wonder Liberal Vision were so worried last September.

You’re absolutely right that there will be a large number of LibDem seats that fall to us (I would put the number nearer 20 than 30, but the minumum will be about 15 pickups).
You overstate the case re the LibDem vote share fall. The LibDems polled 22% in the last general election, and are polling 18-19% now (pre-expenses scandal). But the LibDems have (all the way back to 1992, which is as far as I’ve looked) typically picked up 3-5% in election campaigns – firstly, because they rarely get much coverage outside of elections; secondly, because the LibDems have tended (although this may not re-occur given our reinvigorated activist base) to have better GOTV operations than peers.
It’s also probably pushing the case to suggest that Edinburgh West and North East Fife are Cob/LibDem marginals – in each case the Libs outpolled us by margins of 2.5-1. (The Libs may well lose one – or even both – of those seats, I just suspect it won’t be to us.)
There are also local factcors at play. Huhne in Eastleigh may well keep his seat. Ed Davey in Kingston, although well down the list, may well lose his. Zac Goldsmith is utterly reviled by activists in Richmond, and may fail to pick up what should be an easy gain. Nick Harvey in Cornwall North will probably benefit from the collapse of UKIP. In other Cornish seats, their disappearance will be a disaster for sitting MP – I wouldn’t be surprised to see the LibDems lose three quarters of their Cornish / West Devon seats.
tt
Hi TT,
Let me take your points one by one. Yes, we are looking at figures given the current poll ratings (slightly lower than you are claiming) but we do also look at figure for the Lib Dem vote more similar to the 2005 result.
We’re not treating the seats of Edinburgh West or NE Fife as “marginal”. Our chart shows that it would require a phenomenal swing for the Conservatives to gain them. They appear in our list because they are Con / Lib battlegrounds according to our definition. If the Tories won those two seats then the Lib Dems would be looking at a complete slaughter at the next election.
Yes, local factors are at play (see for example Norfolk North at the 2005 election) but we have already stated that. Forecast UK makes macro level forecasts so while we expect that our individual predictions for specific constituencies might be wrong, overall we come close to the aggregate picture (as our predictions for Wales in 2007 demonstrate).
I wish I could be so optimistic!
We’re currently polling in the low 40s, as opposed to the 32% we got in ’05. Let us simply assume we get 43% at the general election – all from Labour – and the LabDems score exactly what they got in ’05. (If I’m wrong with these forecasts, I suspect it will be because Labour has completely collapsed, and then I suspect we’d see us at 46% and the Libs in the mid 20s.)
On my forecast, we’d see 4 Con voters for every 3 we had previously, and a basically unchanged LibDem vote. This would net us 21 gains on your chart. Now, the real number may be a little higher (I wouldn’t be surprised if Cornwall SE and Truro went our way, and have already made my Kingston forecast.)
But your mistake is to forecast a “swing” – truth is, we’ve gained votes from Labour and the Libs have held it steady. Surely it makes more sense to just multiply our vote by 1.33, and keep the Libs steady.) On your numbers, we’d nab Twickenham. And that isn’t (alas) going to happen. On the ground here, I see all the effort being ploughed into Richmond and (thanks to a great candidate) Kingston.
I think not….
Quite the reverse.
Granted, the situation is now slightly different n’est ce pas?
Hind sight is a wonderful thing, agree with George, life is very different now !!!!
Would be very interesting to see what the results show now with the Clegg bounce still ongoing. Poll of Polls show that Lib Dems could end up with as many as 100/110 M.P.’s on 7th May, Question is ????
Who will be the loser – Tories or labour ???
After the last TV debate, YouGov asked: “How would you vote on May 6 if you thought the Liberal Democrats had a significant chance of winning the election”. The responses: Lib Dem 49%, Conservative 25%, Labour 19%. If this actually happened there would be 548 Lib Dem MPs, 41 Labour MPs and just 25 Tories.
The message here seems to be, vote for what you really want to happen, and it will.
Do minor differences in the polls really matter. They all show that we are heading for a hung parliament with the Tory’s being the largest party in terms of seats won. The question is quite simple Who will LibDem get into bed with to form a government. This will all come down to whether anyone will offer Electoral Reform to them as a aweetener.
They have to go with Cons – Clegg can’t possibly talk about “fair” electoral reform on one hand AND get into bed with the labour party as PM with least # of votes on the other.
Btw it’s neck and neck still – both Cons and LibDem need to improve their share of the vote by 5% in the last week to get a majority, which, based on previous elections and pollsters, doesn’t happen. Interesting that Cons still need to work harder than Labour to get to their majority. Also maybe it’s too late, the people with a postal vote may have already voted Clegg.
Hi MB – You have hit the $64 question right on the head ! IT IS going to be all about P.R. – not transferrable voting etc…. but P.R.
And that will change british politics for ever.