Paula Surridge


A key story for the 2019 election was whether voters who were at odds with the party they voted for in 2017 over Brexit would switch to a party more in-line with their views.

This is a first look at how the EU referendum ‘identities’ voters held related to voting in the election. It considers voters in England only as the picture in Scotland and Wales is very different but the sample sizes in the data make separate analysis more problematic. Scottish and Welsh election studies will be available in the coming months.

The measure of EU identity used here comes from wave 16 of the British Election Study Internet Panel. Which asked people not only if they identified with a ‘side’ on the EU referendum but also, critically, how strongly they did so.


Image for post

These have been combined into a 7-point scale for the purposes of this analysis ranging from ‘Very strongly remain’ to ‘Very strong leave’.


Image for post

Around half of the English respondents were clustered in the two strongest identity categories, with around 17% in the two ‘fairly strong’ categories and less than 10% with no identity at all. It is a remarkably symmetrical distribution across the remain and leave sides.

Looking at how voters in each part of the distribution voted immediately highlights a key story of the 2019 election.


Image for post

The concentration of the leave vote for the Conservatives is much greater than the concentration of the remain vote for Labour. The Liberal Democrats in England managed to out-poll the Conservatives in the two most-remain identifying groups, but only just among the ‘fairly strong’ remainers. While the Conservatives were able to reach across the remain groups with 'fairly' or 'not very strong' identities better than Labour were able to among the equivalent leave groups. This also highlights the importance of the strength of Brexit identities as well as their direction.

Of particular interest is how voters who were not ‘aligned’ with their party in 2017 voted in 2019. To examine this we break down the Labour 2017 and Conservative 2017 voters in the same way as above. It should be kept in mind, of course, that voters for the Conservatives in 2017 skewed towards the ‘leave’ end of the distribution, while Labour’s 2017 vote was more concentrated among the remain groups.


Image for post

As we would expect across the groups the most common vote in 2019 for those who voted Conservative in 2017 is again Conservative. This is overwhelmingly the case from the ‘Not very strong remain’ category through to the ‘Very strong leave’ category. As suggested before the vote, for those voters with very strong remain identity who had voted Conservative in 2017 there was a tension with the ever ‘hardening Brexit position of the party and among this group the LibDems came out on top (though only by a small margin).

The pattern for Labour is a little different. While the Labour party held on to most of those who had voted for the party in 2017 among the remain identifiers, it is a little less successful than the Conservatives were among the leave identifying groups. The ‘drop off’ in Labour support is also more rapid, and for the ‘very strong’ leave identifiers who voted Labour in 2017 only 30% did so again in 2019. This was especially damaging for the Labour party because the main beneficiary here are the Conservatives (whereas among Conservative 2017 remain identifiers the LibDems were the main gainers).


Image for post

These two charts also show that the LibDems were more successful at winning the votes of remain identifiers from the Conservatives than from Labour.

Further more detailed analyses are needed. It is clear that Brexit identity and conflict with party position mattered in the 2019 election, but critically it depends on the strength of that identity. It also shows the dilemma Labour were caught in. There were far more of their 2017 voters to lose from the ‘remain’ identifying groups but the loss of voters in the ‘leave’ identifying groups directly to the Conservatives was especially costly.


Author biography

Paula Surridge is a Senior Lecturer at the University of Bristol. She tweets at @p_surridge. This article was first published on the blog platform Medium and has been reposted with the permission of the author. Image credit: Number 10/Flickr.