YouGov data analysed for this Review by Datapraxis confirms that the three main factors in Labour’s losses were:
To go into an election with any one of these vulnerabilities – an unpopular leader, a problematic position on the key issue of the day, or serious questions about the deliverability of key policies – would be a challenge, one that might be offset by strengths on the other two fronts. But to be in a weak position on all three at once arguably had a “snowballing” effect that was fatal to Labour’s chances of securing sufficient support to win.
Behind these issues lie differing overall perceptions and evaluations of the Labour Party across the electorate, and significant difference in values and outlooks among key groups.
As we saw in Chapter 2, these differences and divisions have been germinating over a long period. However, Chapter 3 focuses on the immediate factors behind Labour’s 2019 loss.
In 2017 Jeremy Corbyn succeeded in winning over many more voters than some had expected - improving both Labour’s and his own personal poll ratings dramatically over the course of the campaign.
Evidence we reviewed from the British Election Study, Datapraxis analysis of YouGov data, and other independent sources indicates that this was not repeated in 2019, and that negative feelings about Jeremy Corbyn had become a key deterrent for much of Labour’s voter base.
According to the British Election Study, in 2017 the difference in voter “likes” between Jeremy Corbyn and Theresa May had narrowed to within the margin of error at the close of the campaign.
Had voters’ feelings about Jeremy Corbyn remained at the peak they reached after the 2017 election, the British Election Study team estimates that Labour’s vote share in this election would have been six points higher - over 38 per cent.
However, Datapraxis analysis of YouGov data shows that perceptions of Labour’s leader began to decline from this point, independently of the Party’s ratings, with a particular drop in Spring 2018. From the end of 2018 views of Jeremy Corbyn declined in step with the Party’s poll ratings.
The timing of these shifts suggests that negative perceptions of Labour’s leader were bound up with people’s awareness of divisions and disunity within the Party.
The sharp collapse in support both for Jeremy Corbyn and Labour between December 2018 and June 2019 coincided with the defection of MPs to form “The Independent Group”, disagreement over Labour’s position on Brexit going into the European elections and the controversy of the Party’s handling of anti-Semitism. (Among 2017 Labour voters who withdrew their support in 2019, the perception that the Party was divided was given as the third most important reason why Labour lost the election, after “Brexit dominated the election” and “Jeremy Corbyn was not an appealing leader”.)
By September 2019 Jeremy Corbyn’s ratings were at record lows. YouGov data shows that 67 per cent of all voters “disliked” him, most of them strongly, and only 12 per cent “liked” him. Even of those who had voted Labour in 2017, more “disliked” him than “liked” him, and most saw him as weak and indecisive.
YouGov data also shows that, between the 2017 election and the eve of the 2019 campaign, both Leave and Remain voters became significantly more likely to regard Jeremy Corbyn as “weak”, and less likely to regard him as “in touch with ordinary people”.
Unlike in 2017, the 2019 campaign itself did little to rectify this position. YouGov data shows some recovery in Jeremy Corbyn’s “favourability” ratings in the last few weeks before polling day. However, according to measures used by the British Election Study, Jeremy Corbyn “gained little ground” during the campaign, having started at a much lower base than he did in 2017.
Labour’s disadvantage on leadership was also reflected in unfavourable “best PM” ratings, despite a late dip in Boris Johnson’s rating.
Semantic analysis of responses to open-ended questions conducted for this Review by Datapraxis suggests that negative views of Jeremy Corbyn were a key factor behind support for the Conservatives.
Among voters who switched from Labour to the Conservatives, concern about Jeremy Corbyn was intense, whichever way they voted in the referendum. According to analysis by Datapraxis, when asked for their thoughts about Labour and its policies, Labour Leave voters who switched to the Conservatives were likely to talk about terrorism, anti-Semitism, what they saw as extreme far-left policies, or unaffordability. A typical quote, flagged up by the semantic analysis as illustrating this segment’s viewpoint, was as follows:
Labour Remainers who switched to the Conservatives also typically expressed negative views of the leadership:
Among the 2017 Labour supporters who didn’t vote in the 2016 referendum and abstained in this election, over three-quarters had negative views of Jeremy Corbyn by 2019. A typical quote which was flagged up by the semantic analysis as illustrating this segments’ viewpoint was as follows:
Even of the 2017 voters Labour hung onto a large minority had negative views of Jeremy Corbyn, suggesting they voted despite, rather than because of, the leadership. In Datapraxis message-testing at the start of the 2019 campaign, among those who turned out for Labour in both 2017 and 2019, 33 per cent of Remain voters, and 45 per cent of Labour Leave voters had negative views of Jeremy Corbyn. Even when we turn to the 1.8 million mostly younger 2017 non-voters who turned out for Labour in 2019, just 27 per cent “liked” Jeremy Corbyn and 35 cent actually “disliked” him; 39 per cent were neutral towards him.
According to the British Election Study, compared to 2017, there was little change in voters’ positions on the EU or “left/right” scales, but there was a drop in how much people liked Labour in general, and a much larger drop in how much they liked Jeremy Corbyn. This suggests perceptions of leadership are a central factor in the change in Labour’s vote share between 2017 and 2019.
British Elections Study’s preliminary conclusion from this evidence is that “the most dramatic difference since 2017 was in Jeremy Corbyn’s likeability ratings and this could account for a substantial proportion of Labour’s drop in support.”
Compared to 2017, Labour lost voters on both sides of the referendum debate. In net terms Labour’s vote in 2019 comprised:
These net figures account both for 2017 Labour supporters who didn’t vote Labour in 2019, and smaller numbers of voters on both sides of the divide who hadn’t voted Labour in 2017 but did in 2019, offsetting the loss of 2017 supporters.
Focusing only on those who had voted Labour in 2017, analysis conducted for this Review by Datapraxis concluded that Labour lost:
Because most of Labour’s 2017 voters voted for Remain, these figures reflect a far higher loss rate for Leave voters than Remain voters. 2017 Labour supporters who had voted Remain were over 75 per cent likely to stick with the Party; those who had voted Leave were less than 50 per cent likely. Those who hadn’t voted in the 2016 referendum were less than 67 per cent likely to be retained.
In the 40 previously held Labour seats lost by the largest margin at this election, Labour’s greatest losses were among Leave voters, accounting for slightly more lost votes than Remain voters and Referendum non-voters put together. In 20 more narrowly lost seats, Labour lost more Remain voters than Leave voters.
One of the most predictive factors for voters Labour lost was how they voted in the 2019 European Parliament election, which effectively served for many as a conveyor belt to both Leave and Remain rivals. 2017 Labour supporters who voted Green, Plaid Cymru, Liberal Democrat or Change UK - The Independent Group in the European Elections were between 65 per cent and 80 per cent likely to return to Labour in December 2019; those who had voted Conservative, UKIP, Brexit Party, or SNP in June were less than 30 per cent likely to return.
British Election Study data indicates that Labour lost 32 per cent of its previous Leave voters to the Conservatives, and another 4 per cent to the Brexit Party; at the same time as losing 10 per cent of its (larger number of) Remain voters to the Liberal Democrats, and another 3 per cent to the Conservatives.
Datapraxis found that Brexit was a major driver of Labour Leaver voters being lost to the Conservatives and the Brexit Party. The overwhelming majority of Labour Leave voters lost to the Conservatives and Brexit Party also preferred a hard Brexit, wanted tougher immigration controls, strongly disliked Jeremy Corbyn and did not trust Labour to lead the country.
Polling data analysed by Datapraxis suggests that those who stuck with Labour through 2017 and 2019 did so not out of enthusiasm for Labour’s second referendum position - those who had voted Remain tended to prefer “stop Brexit” to “final say”; those who had voted Leave were seriously tempted by “get Brexit done” - but because wider policies were more important to them than Brexit.
However, there are clear indications that Labour’s protracted avoidance of a clearer stance on Brexit contributed to perceptions of Jeremy Corbyn’s leadership as weak or indecisive.
Other published polling evidence is consistent with this picture. One poll conducted after the election found that:
Datapraxis analysis of YouGov data found that Labour’s policies were individually popular, but the overall package raised concerns about affordability, economic risks or damage, a lack of realism and crucially, an inability to deliver on the promise of real change.
In open-ended questions, Labour’s bold policies were frequently cited as a positive reason to vote for the Party, and the overwhelming majority of the policies attracted super-majority support - including some of the most controversial such as Inclusive Ownership Funds and free broadband (though some lost voters also cited this as an example of Labour’s unrealistic or excessively “radical” ideas).
The resistance came mostly as people evaluated the overall package of proposals. Some were concerned about affordability, others about negative impacts on the economy or their own personal finance. Many came to the conclusion that the manifesto as a whole was unrealistic, risky and unlikely to be delivered. This undermined the positive response to individual policies, making them seem less credible.
This was a clear contrast to the 2017 manifesto which, while bold, was seen as more deliverable.
Negative perceptions of the Party and its leadership reinforced these concerns; this made it easier to paint the manifesto as impractical and unrealistic. Most importantly, most people simply did not believe that a Labour-led government would in fact deliver real change. 
The impression that Labour’s policies added up to less than the sum of their parts is reinforced by other published polling and focus group evidence. Polls conducted shortly before the election found that:
However, support for individual policies falls when they are attached to the Labour Party.
Another poll found that, when asked about Labour’s plans to “spend significantly more money than the UK has witnessed in our lifetime” in areas like the NHS, schools and nationalisation found that:
The Ashcroft poll carried out immediately after the election found that, among voters Labour lost:
Other polling indicated that:
There is evidence that perceptions of Labour’s manifesto were compounded by the effect of having had a recent Party Conference featuring a number of high-profile policy debates. Among Labour Together’s 11,000 survey respondents, 36 per cent of whom had actively campaigned in the election, Labour’s position and plans around private schools were raised as a problematic issue on the doorstep, despite having not been a particularly prominent element of Labour’s manifesto or campaign. It seems likely that perceptions of Labour’s position on the issue were influenced by misconceptions of what the policy actually was: YouGov polling found 50 per per cent of voters, including 35 per cent of Labour voters, were against “banning” private schools (which was not Labour’s policy). This contrasts with strong support, for example, of Labour’s policy of putting VAT on private school fees.
In Scotland the unpopularity of Labour’s leader was linked to his perceived equivocation on Brexit, which antagonised younger, more city-based voters; and his political reputation, which alienated more traditional, especially older, Labour voters. Apparent disagreements around Labour’s position on a second independence referendum added to the confused message around key election issues.
A pre-election poll found Brexit/Europe to be a key election issue for 56 per cent of voters - ahead of all others. The SNP’s clear position on Brexit allowed them to win the support of many Remainers who had opposed independence, and win back supporters they had lost to Labour in 2017. The SNP’s shift in messaging during the campaign, from focussing heavily on independence to focussing on opposing Brexit, showed that they knew these were the voters who needed to be won over.
Criticism that Labour was seeking to avoid the Brexit question went hand-in-hand with Labour’s confused position on a second independence referendum, and the misplaced belief that a pro-second referendum stance would win support back from SNP voters. Scottish Labour lost pro-UK and pro-independence Remain voters while the SNP continued to win support from Leave voting independence supporters.
In both the independence and Brexit debates, Labour was caught between more forceful voices on either side with a message that, to many, looked confused or evasive.
Scottish Labour’s position was appealing to a tiny proportion of the electorate. A pre-election poll found that just 12 per cent of Scottish voters said neither Brexit nor independence would influence which party they would vote for.
In 2017, Scottish Labour had taken a clear position against independence and a second independence referendum, while also advancing an argument against the SNP’s stewardship of public services. This had succeeded in winning over SNP voters sceptical of the SNP’s lack of vision, and those who were opposed to independence and another referendum.
Differing views of Labour’s leader, Brexit position, and credibility expressed deeper differences across the electorate, such as:
Authoritarian/liberal divides explain some of what happened in this election.
To bring out deeper differences of value and outlook, Datapraxis analysed responses to statements of value and outlook, to see which provided the strongest guide to voters’ preferences and decisions.
Among Labour’s 2017 voters, one of the clearest responses was to the Old Testament quote “an eye for an eye, and a tooth for a tooth”. Of those who agreed with this, Labour lost 44 per cent in 2019, compared to 24 per cent of those who disagreed. 60 per cent of lost Labour Leave voters agreed with it. Of Labour’s 2019 voters, 32 per cent agreed with it.
Another key driver of previous Labour voters from the Party was deep distrust of politics.
Of Labour’s 2017 voters, 41 per cent of those who agreed with the statement that “there is very little difference between the three main parties” were lost in 2019, against just 22 per cent of those who disagreed; 65 per cent of lost Labour Leave voters agreed with this. Given the clear differences between the main parties’ manifestos in this election, this view is more likely to be an expression of distrust, fatalism and anti-political sentiment than an actual assessment of policy.
People’s perception of their own political positioning was also a key predictor of which voters Labour lost. Labour was likely to hold onto voters who identified as “very” or “fairly” “left wing”, but voters who identified as politically in the “centre” were 47 per cent likely to be lost, and those who identified as “slightly left-of-centre” were 32 per cent likely to be lost.
Complimentary analysis by the British Election Study shows that voters are most likely to see themselves as near the “centre”, but in this election saw Labour as much more to the “left” than in previous elections. Voters were less likely to vote Labour the more they saw it as being to their own “left”.
It should be borne in mind that this is a matter of where a party is generally perceived to be on an imagined “left/right” spectrum, rather than a function of its actual policies - on economic policy, at least, academic analysis of survey evidence has concluded that the “centre ground” moved significantly to the “left” after 2010, though in 2015 (the latest year for which this analysis is available) was still to the right of where it had been in 1997.
Using polling data and responses to value statements such as those cited above, Datapraxis have clustered the electorate into 14 groups according to their worldviews, value systems and political tendencies.
This analysis confirms that these fault-lines became fractures that split apart Labour’s voter base in 2019.
Datapraxis’s analysis shows that in 2019 Labour’s most resilient support was found among one group: “The Green Left”, who make up 7 per cent of the electorate. Labour slightly increased its support among “Older Establishment Liberals” - the only group among which Labour increased its vote share.
However, Labour lost large numbers of votes among three groups that in 2017 were a core part of its vote: “Progressive Cosmopolitans”, 9 per cent of the electorate with strong pro-European views; “Centre-Left Pragmatists”, 7 per cent of the electorate who tend to have socially liberal views but somewhat negative views of big business and party politics; and “Anti-Tory Heartlands”, a group representing 6 per cent of the electorate who also tend to have negative views of politicians and a more socially conservative outlook.
Labour’s result was further damaged by significant falls in support among “swing” groups, whose votes tend to be distributed across a wide range of Parties and had played an important part in Labour’s overall vote share in 2017: “Older Brexit Swing Voters”, “Anti-Establishment Hard Brexiteers”, “Older Moderate Traditionalists”, “The Older Disillusioned”, and “Pragmatic Tories”. Together these groups contributed around 21 per cent of Labour’s total support in 2017; in 2019 this fell to 9 per cent.
This analysis shows that the coalition of voters Labour managed to pull together in 2017 - which itself fell far short of what was needed to win a parliamentary majority - crumbled in 2019.
The most immediate pressures producing this collapse were Labour’s worsening difficulties with concerns about the leadership, Brexit, credibility and the underlying divisions in value and attitude these revealed. However, as we saw in Chapter 2, the underlying social and economic forces driving this disintegration go back much further.