What on earth is happening in the opinion polls?

This election was supposed to be in the bag for the Conservatives; the only area of doubt was the actual size of their enormous parliamentary majority. But in the final weeks of the campaign, the overwhelming Conservative leads in the polls have shrunk and even, in a few cases, reached single figures.

Taking the first ten polls of the current election campaign (i.e. sampled from 18 April 2017 onwards) and the most recent ten, what differences do we find in support for the Conservatives and Labour?

On this evidence, the narrowing of the Conservative lead does not appear to be the result of a collapse in that party’s support but rather the consequence of increased support for Labour.

Battle of the ages

Where is that additional Labour support coming from? Given the utter shambles surrounding Conservative manifesto pledges/unpledges on social care, we might reasonably consider a likely source to be the group most affected, namely those aged 65+, who turn out to vote in such large numbers. To test this, I have taken the level of support for Conservative and Labour in the first campaign polls of a number of companies and then compared these figures with those in their most recent post-Conservative manifesto polls. The table below sets out the results:

 Earliest  Poll                            Latest  Poll

* 55+

Allowing for margins of error, there may be some movement to Labour among the 65+ age group during this campaign but it is barely perceptible.

But there is a very different story when it comes to younger voters, and if we repeat the table above using the voting intentions of 18-24 year olds the following picture emerges:

 Earliest  Poll                            Latest  Poll

* 18-34

Here we really do see a substantial shift towards Labour since the start of the campaign. However, this extra support comes from an age group that has the worst record of turnout in elections. One estimate suggested that barely 43% of them voted in the 2015 general election.

Mirror, mirror on the wall, who’s the flakiest of them all?

A MORI poll (1,053 respondents sampled 15-17 May, 2017) sounded a warning bell in their commentary on the survey: “There is a softness in the Labour support, however, as the new poll reveals a clear party distinction when it comes to strength of support – 77% of Conservative supporters say they’ve definitely decided who they’ll vote for (22% may change their mind) compared with 57% of Labour supporters (42% may change their mind). Worryingly for Jeremy Corbyn, two in five (38%) Labour voters who may change their mind say they would consider voting for the Conservative party”.

A ComRes poll (1,012 respondents sampled on 11 May, 2017) found 31% of Labour voters stating that: “I don’t like Jeremy Corbyn but I like the Labour Party” compared with 6% of Conservative voters who said: “I don’t like Theresa May but I like the Conservative Party”. These figures may have changed since the poll was taken but it still leaves Labour voters with a dilemma that is not shared by many who intend to vote Conservative. Finally, there appears to be some evidence that Labour is relying more heavily than any other party on support from people who did not vote in 2015 saying they will vote Labour in 2017. If they do vote, then their prior non-voting history will not matter a damn; but people tend to be serial non-voters and support from this group does not offer the firmest foundation for victory.

Further reasons to be careful

During the 2015 election campaign some brave voices (mine not among them) kept querying the relentless talk about a hung parliament by drawing attention to key poll findings beyond bare voting intention figures; these did not support the many polls that put Labour narrowly ahead. How are these key elements in public opinion performing in 2017 as we approach polling day? How have these changed during the campaign, if at all, and if they have, do they underpin the current narrative of a much closer election result than was predicted at the start of the campaign?

Party Leaders

Most campaign polls carry questions about the party leaders. I have chosen the Opinium series to register movement throughout the campaign to date.

On this measure, Mrs May has moved from a net approval rating of +21 points at the start of the campaign to one of +11 points in the company’s latest poll. By contrast, Mr Corbyn has moved from -35 points at the start to -11 points now. The improvement in Mr Corbyn’s ratings will clearly be welcome to his party but they are still left with the difficulty that an initial gap of 56 points between him and Mrs May is now ‘only’ one of 22 points.

Best to manage the economy

Here, I have taken an ICM question and contrast the answers from the start of the campaign with their latest findings

Q Irrespective of which party you yourself support, which team do you think is better able to manage the economy properly?

Once again, we see positive improvements in Labour’s standing on this important indicator but the 39-point Conservative advantage at the outset still remains one of 20 points. And just one-in-five people (including 58% of Labour voters) thought Labour could handle the economy better than the Conservatives.

Conclusion

For a variety of reasons, I find this election confusing. I agree with a distinguished colleague who observed that many people regard it as “unnecessary – not unimportant – but unnecessary”. I believe we will struggle to match the 2015 general election turnout of 66.2%. If that is so, then as others have commented, which groups of electors show a greater determination to vote will be crucial to the outcome: they always are but which ones will they be in 2017?

The fact that bookmakers, on seeing me, break into broad smiles rather than cower in fear behind their counters says all you need to know about my unique insights in forecasting election winners and losers. But all that I, indeed, all that any of us can do is review the evidence.

The Conservative vote has not plummeted, despite their astonishing blunders throughout the campaign. The most recent uncomfortable polls for the Conservatives still show the bulk of 2015 UKIP voters switching to them.

The Lib Dem “no show” in this campaign suggests no impending slaughter of Conservatives who won seats from them in 2015. Scotland treads its own path but there is still the prospect of a few Conservative gains there.

In Wales, the YouGov polls during the campaign have ricocheted like a ball in a squash court but there seems little evidence of any significant Labour advance. Which leaves us, as always, with England, where 82% of Westminster constituencies are being contested.

Outside London and some of the major conurbations, Labour has suffered attrition in England for over a decade: not in Newham but in Nuneaton; not in Islington but in Ipswich; and not in Brent but in Bedford. The co-existence between socially-conservative, white working class supporters and socially-liberal metropolitans within the Labour Party, that underpinned its past election victories, has, for many, ended in bitter separation: they found no space left for their views or their values in Labour’s public square. Has Jeremy Corbyn reversed all that in a six-week campaign?

Perhaps: but there are legitimate grounds for scepticism. Labour support among young voters may hold the key to this election but it is surely troubling that BMG Research (1,374 respondents sampled 21-24 April, 2017) at the very start of the campaign found almost half of 18-24 year olds were unsure whether they were registered to vote. Such a lack of awareness does not sit easily beside stirring accounts of a galvanised youth vote that is destined to turn the tide at this election.

On key issues such as the economy, leadership, immigration and Brexit the numbers still appear solidly on the side of the Conservatives, despite all the help they insist on giving the Labour campaign.

In the end, I am simply left with the wise counsel of the incomparable Damon Runyon (adapting words from Ecclesiastes, in a way the author might not have totally endorsed):

“The race may not always be to the swift, nor the victory to the strong, but that’s how you bet”.