The Week in Polls 18th-24th March

This week I have included an average at the bottom of the table. I'm am aware of the problems associated with a crude analysis of the mean and so I'd like to state to begin with that it is just there to quickly illustrate the differences between each week's article.

 

Fieldwork	Company	CON	LAB	LD	Swing
23-24 Mar	YouGov	36	34	17	2.5%
22-23 Mar	YouGov	37	33	18	3.5%
21-22 Mar	YouGov	36	32	20	3.5%
19-22 Mar	Mori	35	30	21	4.0%
19-22 Mar	Opinium	37	30	15	5.0%
17-22 Mar	Harris	35	28	17	5.0%
18-19 Mar	YouGov	38	31	19	5.0%
17-18 Mar	ICM	38	32	19	4.5%
17-18 Mar	YouGov	36	32	20	3.5%
10-16 Mar	Harris	36	28	18	5.5%
	Average	36.4	31	18.4	4.2%

 

This week's polling has made very interesting reading. In last week's post I noted that the disparity between some of companies was probably more intriguing than the actual polls. However, this week we have seen a degree of consistency that belies the events that have occurred since last Wednesday. As the table shows, the Harris poll that was conducted before last week's post, but published after it, shows the largest gap. The Conservatives did enjoy a five point lead over the weekend in another Harris poll that was backed up by Opinium. This is almost certainly down to the strikes and then the cash for influence scandal but six months ago you'd have been laughed out of any serious political discussion if you suggested that six weeks before the election the largest leads would be 5% and they'd come from two of the more inexperienced pollsters.

Clearly, Labour are doing very well. So well it almost defies belief. Last Thursday I could only speculate on how Brown's 'mistake' at the Iraq Inquiry and the British Airways Strike would affect the polls but it was more a case of how far the Tories would pull away, not if. Throw in an unexpected scandal and SamCam's pregnancy then the Tories must be wondering what they have to do to pull away! Perhaps Cameron's call for workers to break the picket line had a negative effect as it appeared reckless and in support of big business. Regardless of the rights and wrongs of the dispute Cameron's actions were without question inflammatory move. But the Mori poll showed Cameron easily ahead of Brown in approval ratings with no sudden negative movement for the Tory leader from their last poll a month ago.

Maybe the voters just see the dispute as a service industry strike affecting the upper middle class and most voters don't actually care. Or it might be pushing 'old' Labour voters, who had deserted the party, to back the Government as real ideological differences begin to emerge for the first time in years; well, at least New Labour aren't against the unions. Regardless, pushing the strike doesn't seem to be helping the Conservatives and I imagine they'll try to switch their focus in the coming days; probably to the budget. Cameron's 'spring of discontent' line might begin to work if other industrial action, like the planned Network Rail, happens as this will hurt a larger percentage of the electorate.

Last week I asked for some more specific polling on the marginal constituencies and I've sort of got what I wanted. On Tuesday YouGov published a London poll for which the fieldwork was carried out during the weekend. Their headline figures of CON 40%, LAB 31% represents an 8% swing to the Conservatives in the capital since 2005. This was three points higher than YouGov's national poll carried out at the end of last week. It's worth remembering that given Heathrow's close proximity to London this is where the real affect of the BA Strike is likely to be seen. We'll be looking at the London constituencies at the beginning of next week so I'll wait until then to see how this might impact on the seats tally.

Also, today MORI released a poll of constituencies where Labour are defending majorities between 10-18% against the Conservatives; ie. The seats the Tories need a 5-9% swing in. These seats wouldn't be considered marginal in most normal circumstances but up until recently the Conservatives were easily in reach of these targets. But with the polling trend suggesting the lead nationally is almost certainly below 5% the Tories would have been hoping for some good news from this poll. Unfortunately for them the topline figures were CON 37%, LAB 41% which represents only a 5% swing. This is a point up on MORI's national poll which was conducted at the same time. On this evidence the Conservatives are going to fall short of a majority and we'll all get to see how the parties deal with the realities of a hung Parliament.

Having said all that, the elephant in this blog post is yesterday's budget. Tonight's YouGov poll will be the first to measure public opinion since Darling's speech and we'll have a better idea what the electorate actually think, as opposed to what the media think we think! Evidence suggests that there will be a budget slump, not a bounce, for Labour and this week might be as good as it gets. But the markets seem happy with the content and Labour's 'don't make any sudden moves' approach to the economic recovery is showing signs of success. They might manage to stay at the same polling levels.

The Conservatives traditionally do better than the polls suggest, but on this evidence this is still worse than last week when they averaged 38%. The Tories need to find something to turn this around quick as even a swing of 5% isn't good enough. With little time left we may well be witnessing one of the greatest political comebacks in British history! Or it could just be another 1992…!

Finally, I apologise to the Liberal Democrats for virtually ignoring them this week. Sadly for them this is a reflection of how the MSM has treated them and is part of the reason for them slipping under 20% again. Also, the Dolphin House Four can't have helped their party's image but they do have the Chancellor Debate on Monday and thus an excellent opportunity to show the voters that they are a credible alternative to Labour. Cable should beat Osborne hands down and I'm sure Liberal Democrats up and down the country are eagerly awaiting this match up. I still expect them poll close to their 2005 levels but the issue will be where this vote is concentrated.

The Week in Polls (11-18 March)

This is the first of a weekly series on the state of the polls. As much as we are committed to looking at the election at a local level, the national picture can not be disregarded. And as the usual arguments against polling begin to surface, they are by far the best gauge of nationwide opinion and they are also, rightly or wrongly, heavily reported on in the MSM. The plan is to bring you the polls published during the last week with a bit of analysis on the way this election is panning out. The 'swing' column relates to the LAB>CON swing and can be compared to our constituency analysis in LAB/CON marginals and is based on rounded 2005 figures of CON - 33%, LAB - 36% and LD - 23%.

Fieldwork	Company	CON	LAB	LD	Swing
16-17 Mar	YouGov	36	32	20	3.5%
15-16 Mar	Angus Reid	39	26	21	8%
15-16 Mar	YouGov	37	32	19	4%
14-15 Mar	YouGov	37	32	21	4%
12-15 Mar	Opinium	39	28	16	7%
12-14 Mar	ICM	40	31	20	6%
12-13 Mar	YouGov	37	33	17	3.5%
10-11 Mar	YouGov	37	34	17	3%
10-11 Mar	ICM	38	31	21	5%
9-10 Mar	Angus Reid	39	26	18	8%
4-10 Mar	RNB	39	31	20	5.5%

With the election seven weeks away there is still al long way to go but the polling well underway. The number of companies surveying this election is already in double figures and the fact the Indian based RNB published their first poll this week shows that there is still time for more to enter the fray. The most striking trend in the numbers at the moment is the disparity between some of the pollsters. YouGov didn't report a Conservative lead larger than 5 points this week yet another internet panel pollster, Angus Reid, published two polls with a 13% margin for the Tories. Both can't be right and if this trend continues then it won't just be the candidates in marginals with squeaky bums on election night.

ICM has generally been considered the most consistently accurate company in recent years. Their two polls this week have shown a 7 and 9 point lead for the Conservatives, which puts their forecast in the middle of the two internet pollsters. As such, it's probably safe to conclude ICM are more likely to be in the 'right' area when it comes to the headline figures.

As for how all this looks for the three main parties, I suspect the Lib Dems are the happiest followed by Labour. Although Labour seemed to have hit a glass ceiling in the low thirties this is a lot better than their figures at the end of last year. If they do in fact keep the swing lower than 6%, as YouGov seems to suggest, I imagine they'll be pretty pleased. The BA Strike hit the news at the end of last week but despite the constant coverage this hasn't really corresponded in a significant shift towards the Conservatives. In fact they appear to have lost support and any gains that Cameron may have made from yesterday's PMQs performance look set to be wiped out by fresh Ashcroft allegations dominating the news cycle today. The Lib Dems, on the other hand, have enjoyed a good start to this week which is likely to be due to their spring conference at the weekend. They appear to have finally broken through the 20% barrier they've been stuck under for most of the year and it goes to show that when the third party receive some media attention they get a bounce in the polls. This will defiantly please the party in the run up to the campaign, especially as they now have the Chancellors' debate to show off their prize asset, Vince Cable, to the nation.

At the moment an average of the figures would put the Conservatives in the high thirties, Labour in the low thirties and the Liberal Democrats around twenty. I think the Lib Dems will rise slightly from this when the campaign gets fully into swing but the two main parties seem to be near their limitations. Labour will surely lose at least 10% of its vote from 2005 and this would leave them hovering just above the 30 mark. As for the Conservatives, it has been a while since they've been consistently above 40%. As the RNB poll shows, 39/31 is only a 5.5% swing which would leave them a couple dozen seats short of a majority on a universal swing. What would be nice is to see some more marginal polls, like this one from Populus, to see how the main parties are fairing in the key seats. Last week it seemed the Tories were heading for a small majority but a week is a long time in politics! Perhaps the next seven days will bring more marginal polls and we’ll be able to see the state of play where this election will be won and lost.