Archive for the 'Labour' Category


Miliband’s Achilles’ heel: those who backed Brown

Saturday, August 16th, 2014

Never mind the LD switchers, the biggest threat to Labour was already in the Red column

One assertion that receives a regular hearing on politicalbetting is that Labour is in an extremely strong position to win the next election thanks to that group of voters who switched from Lib Dem to Labour in 2010.  They’ve been consistent in their support ever since and remain favourably disposed towards Miliband and Labour.  Add in that UKIP’s support has come disproportionately from Con, that Labour’s vote is more efficiently distributed, and that precious little direct Con-Lab has taken place (meaning it’s unlikely much could swing back), and it’s easy to see why many can picture Miliband on the steps of No 10 next May.

All of which is true but it’s still not the whole picture.  The weak spot in Labour’s coalition is not those who’ve joined since 2010 – those who’d traditionally be seen as swing voters – it’s those who were already there.

That may seem remarkable given that Labour polled their second-worst total since WWII in 2010, only just better than 1983.  You would think that they’d have been somewhere near their base with a score like that.  Yet history suggests it’s entirely possible for a party to go backwards from a defeat: the Tories managed it in 2001 as did Labour in the aforementioned 1983.

And the risk is real enough: the August Mori poll reported Miliband’s overall net satisfaction rating as -29%, in the same range as Hague and Duncan Smith when they were Leader of the Opposition.  Even more concerning for him, some 41% of those saying they’d vote Labour were dissatisfied with him.  It’s true that Cameron’s net rating with Labour supporters is -52% but then you’d expect supporters of one party to give poor scores to leaders of the others, especially when they rate their own so badly.  Obviously, we shouldn’t read too much into one or two subsamples from just one poll – but this isn’t just one poll: YouGov’s results from 7-8 August, for example, had similar findings.

Even more notable is that Miliband’s rating among 2010 Labour voters with Mori was -5%: it’s only the LD switchers that pull his score into positive territory.  YouGov recorded a small positive balance but still three in seven of Labour’s 2010 support reported dissatisfaction.

Does this matter?  Won’t those same core Labour voters, with their even lower opinion of Cameron, turn out even if unenthusiastically?  Perhaps, but we shouldn’t bank on it.  It is, after all, easy to say you’ll vote when responding over the phone or monitor.  Going out and actually doing it in person (or even by post), is another thing: hence the disparity between how sure people say they are to vote and how many end up doing so (the Mori poll referred to earlier found 78% rating their likelihood to vote as between 7 and 10 out of 10).

Similarly, while they’re not likely to be tempted by the Tories or Lib Dems, Labour’s coalition of voters includes groups who could find either UKIP or the Greens attractive.  The Mori poll found a loss of 11% of Labour’s 2010 vote to one or other of those two parties; the Tory figure for comparison was 13%.  (YouGov found similar results; ICM, by contrast, reported a loss of just 4% of Labour’s 2010 vote to Green or Purple, excluding Don’t Knows).  One could of course say that offers Labour an opportunity as well as a threat: if these voters could be persuaded to return, it places even less need on Team Miliband winning Tories over.  That’s true, but it’s a big ‘if’.

David Herdson.


There can be no getting round the fact that Tories are still being the most hurt by the UKIP surge

Monday, July 28th, 2014

And a lot of 2010 non-voters seem to back Farage’s party

The above chart is based on the aggregate data from Lord Ashcroft’s latest round of CON-LAB marginals polling which had a total sample of 14,004.

The first factor to stand out is that much more of UKIP’s current support in these key battlegrounds continues to come from ex-Tories than from ex-LAB voters. This means, of course, that the blues will benefit most should UKIP support fade.

Secondly, given UKIP only got 3.1% nationally in 2010, a very high proportion of current UKIP voters did not vote at the last election.

All this unerpins the claims by LAB in the Telegraph this morning that “Ukip voters will make Ed Miliband Prime Minister”. The report quotes a LAB figure:-

“The Tories lose a lot more than we do from a decent Ukip performance,” said a senior Labour campaign source. “The whole election could hang on how many of their current voters stick with them next May.”

I think that’s right and this will impact on Labour’s approach in the coming months. Ed Miliband’s team will be increasingly resistant to pressure from some in the party to make policy moves to attract UKIP votes.

That means, I’d suggest, no Labour promise on an EU in/out referendum.

Mike Smithson

2004-2014: The view from OUTSIDE the Westminster bubble


Why Blairites like John Rentoul have got to stop looking at GE2015 through the prism of 1997

Monday, July 14th, 2014

It’s a totally different election with very different dynamics

There’s no doubt that Tony Blair’s GE1997 victory, coming as it did after four election defeats over the previous 18 years, was a stunning success. Blair did it by reinventing his party so it would appeal to large swaithes of voters who never before had done anything other than vote Tory.

But because that result was so good for the party doesn’t mean that the Blair approach is the only one that will work for the Red team or that it is even possible now. Take this from the Blair biographer and ongoing Blair enthusiast, John Rentoul in yesterday’s Indy on Sunday:-

“My view, and this cannot be based on opinion polls, is that when the voters come to choose they will shy away from the prospect of Miliband as prime minister, just as they shied away from Neil Kinnock in 1992.”

I’d suggest that it is very dangerous to ignore the numbers, as Rentoul is suggesting, and base analysis on gut feelings, anecdote, or previous positions.

In 1997 the Blair challenge was to attract 1992 CON voters. At the coming general election all the polling points to very little switching between 2010 CON and 2010 LAB. The main movement has been the big post-coalition shift of 2010 LD voters to LAB and the biggest priority for the red team is to retain them in the key marginals.

That’s still holding up and notice from the table above how this key segment views Mr. Miliband. This is a view of Ed that has been seen in mega-polls whenever the sub-sample of switchers has been shown.

Would Rentoul’s choice for LAB leader, David Miliband, have had anything like this level of appeal to the voters that matter?

Mike Smithson

Ranked in top 33 most influential over 50s on Twitter


Ex-UKIP leader chosen by Tories to fight the seat’s where Farage is said to be interested

Sunday, July 6th, 2014

The tactical anti-UKIP vote could now go to LAB


Labour in Newark: Ruthless or wrongheaded?

Saturday, June 7th, 2014

Soft-pedalling the campaign is a sign of both weakness and strength

Conventional wisdom says that general elections are won or lost based on the decisions of a few tens of thousands of swing voters across the country’s marginal seats.  As an assertion, it was never entirely true – those voters made next to no difference in 1983 or 1997 for example – but in an increasingly fractured party system, the assumptions on which it rests become more and more questionable.

Those assumptions (and indeed, the whole concept of ‘swing’), go back to a time when there were only two main parties and there were high and stable turnouts.  A lost voter for one party was a gained voter for the other.  While we shouldn’t over-egg the death of Uniform National Swing, it’s far from the psephological rule it once was.  Indeed, perhaps not coincidentally, as campaigning techniques have become ever more sophisticated in targeting, so the extent to which those voters ‘decide’ elections has declined: you can’t expect a uniform swing if you don’t have a uniform campaign.

Which brings us to Newark.  On one level, Labour not campaigning too hard there was understandable: they were starting well back and while the principal party of opposition has won by-elections before on the sort of swing required, not with a poll lead in low- to mid-single figures.  A gain would have been nice for them but it was never on the cards.  There’s an attractive argument that it’d be much better to spend the money saved on the target seats instead.

Not being distracted by the tempting but unlikely opportunities Newark offered is in that sense a sign of strength from the Labour campaign HQ: that they will keep their eyes on the important priority, namely winning the seats they actually need to form a government.

On the other hand, it’s not a strategy without risk.  Soft-pedalling always brings with it the possibility of a much worse result than anticipated as the parties who are going for it squeeze the rest out.  Reports of natural Labour supporters voting Con to keep UKIP out are therefore unsurprising.  As it’s also entirely possible that some voted UKIP to inflict a defeat on the Tories (the desire for such a result having been initially put forward as one reason Labour didn’t try too hard in the first place), one has to question whether the net effect was worth the sacrifice.

That kind of tactical leakage is in microcosm, symptomatic of the bigger problem: which are the target seats?  UKIP are up over ten per cent since 2010 and the Lib Dems down by even more.  With considerable differences in how that will play out across the constituencies, what might be winnable (or losable) becomes a lot harder to call than a simple application of UNS would make it.

The important thing about Newark was that Labour chose not to pursue their claim as chief challengers to the Tories, despite starting in second (and despite audaciously, but falsely, claiming to represent a One Nation tradition).  Newark might have been a challenge too far but if Labour can soft-pedal a by-election, they can and probably will soft-pedal plenty of constituencies at a general election.

That’s fine as long as you pick and choose correctly, though it risks the party atrophying elsewhere.  Get it wrong though – and with the vote-churn there’s been since 2010, it’s far easier to get it wrong – and you could both miss out on makeable gains and, even worse, lose seats previously assumed to be safe.

David Herdson


Locals 2014: Afternoon update – The UKIP fox is in the Westminster hen house

Friday, May 23rd, 2014

Like opinion polls, it is wise not to focus on one or two councils, but look at the broader picture.

Often success  equals performance minus anticipation, using Rallings and Thrasher’s projections for the locals, of Lab 490 gains, the Cons and Lib Dems 220 and 350 losses respectively, and UKIP to make 80 gains, so far it is a great set of elections for UKIP as they have impress results all over England, and for the coalition parties an as expected, and as good as Rallings and Thrasher were predicting for Labour.

This has led to some Labour MPs to go criticise their own side, but Labour has some impressive results,  especially in London, that should be used to calm the nerves.


There has been talk of an electoral pact between the Cons and The Kippers from some Tory MPs, I think the Sun’s political editor has it right.

The other party that maybe happy as well, the Greens, especially on Sunday when the Euros are out.

Finally the bookies take on the results so far.



Polling analysis: The big driver of Labour’s decline has not been been a move to CON but to don’t know

Wednesday, May 14th, 2014

Also Ukip sees big increase in 2010 LAB switchers

The chart shows the responses of those telling ICM in their last six polls that they voted Labour at GE2010 but have now switched to either “Dont know/Refused” or Ukip.

As can be seen there has been a marked increase in switching to the purples but by far and away the biggest element has been the increase in 2010 LAB voters now saying they don’t know or simply refusing to given an answer.

In many ways this is positive news for Ed Miliband and his team. Former voters who are now unsure are clearly good prospects and the hope must be that they’ll return to the fold in May 2015. The reduction in LAB don’t knows in the final few days of the 2010 campaign was a big reason why the party did much better than any of the polls.

But there has been switching to Ukip – from just 1% back in December to 6%. These are small numbers which are part of the sub-set of LAB GE2010 voters so there is a high margin of error.

Mike Smithson

2004-2014: The view from OUTSIDE the Westminster bubble


Labour drops to its lowest level with YouGov since the summer of 2010

Wednesday, May 14th, 2014

The red team’s terrible polling week continues

When on Monday monrning the latest Populus poll came out showing LAB on 36% just a point ahead of the Tories it didn’t attract much attention. Since its big party ID weightings change in February the firm has been showing some of the worst LAB position and this just seemed to follow that pattern.

Then on Monday afternoon the first of Lord Ashcroft’s new weekly phone polls came out with the Tories two points ahead – the first survey of any sort since March 2012 with Labour behind. This got lots of attention but the initial thought was that this was an outlier.

That perception changed three hours later when news of the ICM monthly phone poll for the Guardian emerged showing a similar 2% lead for the Tories. Poll watchers waited to see what the nightly Sun Tweet about its YouGov poll would report at just before 10pm. It had LAB 35 to CON 34.

So last night there was a lot of anticipation for the latest YouGov based on fieldwork that closed yesterday afternoon. The result is featured above – LAB and CON level pegging on 34%. The last time that Labour was that low with the firm was in June 2010 only weeks after the party’s GE2010 defeat.

    So with four pollsters in two days showing the same broad picture the trend is becoming clearer – LAB is down and this is happening only days before what will be the final national electoral test before the general election – the Euro and locals a week tomorrow.

My reading is that EdM and his team are paying the price for not having a clear message for the Euro elections. Their campaign has almost totally avoided any mention of the EU unlike the other three parties competing for votes which each have specific and relevant positions.

Coming up today or tomorrow should be the Ipsos-MORI Political Monitor for May a survey which like Ashcroft and ICM is carried out by phone.

Mike Smithson

2004-2014: The view from OUTSIDE the Westminster bubble