Voter Migration by Group 2017-2019

This page first posted 21 January 2020

This article is part of a small series on voter migration at the 2019 election. See also Voter Migration by Party.

1. Introduction

How did voters change their vote between 2017 and 2019? One way of seeing this is with our migration graphic which shows how voters moved around based on their past vote in 2017.

Another way is to look more deeply at various categories of voters and see how voters with similar characteristics behaved. In particular we will look at three characteristics:

Other characteristics, such as social class, education, religion and geographical region, were not as important as the three main drivers in explaining voter behaviour.

2. Tribe

Electoral Calculus has analysed the results of the 2019 General Election in terms of its seven political "tribes". These groups of like-minded voters have been created by Electoral Calculus as a useful way to understand political opinions and trends.

The Electoral Calculus analysis shows that the Conservatives under Boris Johnson gained relative to Labour with voters in all the tribes which are not right-of-centre economically. But the Conservatives appear to have lost a little ground with voters who are right-of-centre economically.

Labour lost ground in almost all groups.

The 'Traditionalists' tribe of the left-wing working-class was particularly striking. This group had previously been strong supporters of the Labour party, with about three voters in four voting for Labour. But at this election there was a swing to the Conservatives of around 15pc-20pc, which was the largest swing of any the seven tribes.

2.1 Tribe Definitions

The table below shows the seven tribes. They are defined in terms of their political attitudes on three separate axes: economic (from left to right), nationalistic (from globalist to nationalist) and social (from liberal to conservative). There is a high correlation between nationalistic and social attitudes, so socially liberal voters tend to be globalist and more on the Remain side of the EU referendum.

The size of the tribe, relative to the GB electorate, is given as a percentage in the final column.

TribeBrief descriptionPolitical and Demographic CharacteristicsSize
Strong LeftCorbynistasLeft economically, socially liberal; strong Remainers; middle class, older, well-educated4%
TraditionalistsWorking-class LabourLeft economically, socially moderate; mild Leavers; working class, older, more religious10%
ProgressivesBlairitesMildly left economically, socially liberal; more Remainers; middle class, younger, well-educated11%
CentristsMiddle of the roadCentrist; often middle-aged24%
SomewheresWorking-class conservativesSlightly left economically, socially conservative, strongly nationalist; strong Leavers; working class, older, less well-educated12%
Kind Young CapitalistsRight-of-centre liberalsQuite right economically, mildly liberal; mild Remainers; younger, well-educated24%
Strong RightRight-of-centre conservativesRight economically, nationalist, conservative; more Leavers; owner occupiers, Christians15%

The seven political tribes were developed by Electoral Calculus and published in April 2019. Based on British Election Study (BES) polling involving 13,600 people, Electoral Calculus has used modern techniques of cluster analysis to define groups of likeminded voters. Voters' opinions are assessed on three axes: economic, nationalistic and social; and their resulting 3D position is used to allocate them to one of seven political tribes.

More details on the seven tribes at 3D Politics.

2.2 Vote shares

Electoral Calculus has estimated the vote share at the 2019 election for each tribe. Each row shows the estimated vote share across the parties for a particular tribe. The final column 'Swing' has an estimate of the swing in support from Labour to Conservative since 2017.

TribeCONLABLIBBrexitGreenNATSwing
Strong Left4%64%16%0%4%12%8%
Traditionalists23%52%12%3%3%7%18%
Progressives9%56%21%1%5%8%8%
Centrists44%35%12%2%3%4%7%
Somewheres66%20%4%5%1%3%10%
Kind Young Capitalists42%33%16%2%4%3%-4%
Strong Right84%6%5%3%1%1%-5%
Total44%33%12%2%3%4%4%

The Conservatives were the most popular party among 'Somewheres' and 'Strong Right' voters, and the Labour party were most popular among the Strong Left, Traditionalists and Progressives. Among Centrists and 'Kind Young Capitalists' the Conservatives were a bit more popular than Labour.

2.3 Changes since 2017

Electoral Calculus has also estimated the change in support for the Conservative and Labour parties for each tribe between the 2017 and 2019 elections.

The chart shows the change in vote share of each of the three major parties and each tribe.

Change in vote share from 2017 to 2019 by Tribe

Another way of seeing the same data is to show it on a triangular-inspired lattice. This is parameterised on the horizontal axis which is the difference between Conservative and Labour vote shares. The left-hand corner, at −100%, represents the case where Labour gets 100pc of the vote and the Conservatives get nothing. The right-hand corner, at +100%, represents the case where the Conservatives get 100pc of the vote and Labour get nothing. The central point, at 0%, indicates a place where the Conservative and Labour vote shares are equal. The Liberal Democrat vote is on the vertical axis.

The grey diagonal lines indicate the edge of the feasible area, which is a triangle. If the Lib Dems get, say, 40pc then the Con−Lab score must lie between −60pc and 60pc.

The behaviour of a group from 2017 to 2019 is indicated by a line arrow from the group's 2017 position to its 2019 position. A swing from Labour to Conservative would be a horizontal line going from left to right, and a swing to the Liberal Democrats would be a diagonal line at 45 degrees (up-and-right for a swing from Labour, and up-and-left for a swing from the Conservatives).

Swings from 2017 to 2019 by Tribe

The Conservatives have gained a lot of support from voters in the two working-class tribes, 'Traditionalists' and 'Somewheres'. These migrating voters had predominantly voted to Leave in the EU Referendum, so were likely attracted by the Conservatives' slogan of "Get Brexit Done", and were less concerned about traditional left-wing economic issues.

But a consequence of that was a measured reduction in Conservative support among the economically right-wing voters of the 'Kind Young Capitalists' and 'Strong Right' tribes. Many of those who left the Conservatives had voted to Remain in the referendum and switched to the Liberal Democrats, indicating that Brexit was more important than economic issues for these voters as well.

On the Labour side, the 2019 performance was very disappointing. They lost ground in almost every tribe and had no material gains. Their largest loss was from voters in the Traditionalist tribe of left-wing working-class voters. This tribe was evenly split on the EU Referendum, but those that switched to the Conservatives had overwhelmingly voted to Leave in the referendum.

Labour also lost 'Somewhere' voters in a similar way. This tribe was more strongly for Brexit in 2016, and almost all voters who defected from Labour to the Conservatives had voted to Leave. The imperative to 'Get Brexit Done' appears to have resonated very strongly with this group. Across other tribes, Labour Leave voters left Labour to switch to the Conservatives and the Brexit party.

But Remainers were not satisfied with Labour either. Among the Strong Left, Traditionalists and Progressives, many Remain voters switched to the Liberal Democrats and others abstained. This decreased the Labour vote in these tribes which gave the Conservatives a relative advantage there.

Overall, Labour suffered abstentions and losses of three types:

Labour's Brexit triangulation strategy does not appear to have worked well as both Leave and Remain voters left the party to support other parties with more clearly-defined Brexit positions.

3. Age

Another useful perspective is to categorise voters for their age. We use the pollsters' standard age buckets of: 18-24, 25-34, 35-44, 45-54, 55-64, and 65 plus.

The triangle chart below shows the arrows for each age group going from their 2017 position to their 2019 position.

Vote share change from 2017 to 2019 by Age

The main feature is that younger groups are more inclined to Labour and older groups to the Conservatives. In terms of changes, all age groups also moved away from Labour. All the arrows are directed to the right and upwards, showing a general trend away from Labour and towards the Conservatives and Liberal Democrats.

The trends were mostly towards the Conservatives, though the youngest group of 18-24 year olds swung mostly to the Liberal Democrats (the red arrow is going up diagonally, rather than across horizontally). The middle-aged group of 35-44 year olds has the biggest swing to the Conservatives.

4. EU Referendum

Our final viewpoint is that of how people voted at the 2016 referendum on EU membership. Voters are divided into three categories: Leavers, Remainers, and those who did not vote (because either they abstained or were too young in 2016).

The triangle chart below shows the arrows for each EU referendum group going from their 2017 position to their 2019 position.

Vote share change from 2017 to 2019 by EU Referendum vote

Each group behaved very differently. The Leavers, who were already strongly Conservative, became even more Conservative with a swing of nearly 10pc from Labour to Conservative, indicated by the left-to-right blue arrow. 'Get Brexit Done' resonated loudly with this group.

The Remainers were fairly strong for Labour in 2017, but Labour did not benefit further from that. The vertical yellow arrow indicates that both Labour and Conservatives lost remainers to the Liberal Democrats. This equivalence neutralised potential Conservative losses as they did not lose any ground relative to Labour.

Finally, the non-voters in 2016 are represented by the grey arrow pointing upwards and rightwards. This indicates that these voters swung partly to the Conservative and partly to the Liberal Democrats. That suggests that those voters might also have been split in spirit between Leavers and Remainers.

5. Conclusions

Overall Labour lost voters in all directions, especially from the traditional working class. Leavers swung heavily to the Conservatives, and Remainers from both parties swung mildly to the Liberal Democrats. The Conservatives generally gained voters from all categories, except the 'kind young capitalists' which swung to the Lib Dems, as did youthful Labour supporters.

The Conservatives were fortunate that Labour had a voter-repelling combination of poor leadership, hard-left policies, and a triangulated Brexit muddle. All of these could be fixed by a new Labour leadership if it moved towards the centre and if a completed Brexit process defused the EU issue. One worry for the Conservatives might be the defection of the youthful right-wing vote (the KYCs) to the Liberal Democrats. If the Conservatives desert that political space, then it could create a vulnerability for them at the next election.


Return to Psephology home page.