Brexit was hardly mentioned in the 2024 election campaign. That might suggest that voters have forgotten all about the issue that defined the previous election in 2019. But Jon Wheatley shows that the way people voted in 2024 still reflects the 2016 Brexit referendum result.
In Britain’s first post-Brexit general election campaign, the Brexit question was conspicuous by its absence. Despite opinion polls showing that a majority of the population perceive it as a failure, most mainstream politicians did not engage with the issue. This included the Liberal Democrats, erstwhile champions of the Remain cause, who – despite their formal commitment to rejoining the European Union eventually – chose instead to focus on health and social care. Only the Scottish National Party and Plaid Cymru campaigned on the issue, arguing that Brexit had caused untold damage to the Scottish and Welsh economies. Labour, which had reluctantly agreed on the need for second referendum in 2019, now steadfastly opposed rejoining not only the EU, but also the single market and the customs union.
At first glance, the election results suggested that Brexit was no longer an issue for voters either. Labour retook the so-called “red wall” seats, areas of northern England that had voted for Brexit in 2016 and had then swung strongly to Boris Johnson’s Conservatives in 2019. Similarly, the Liberal Democrats successfully retook swathes of Leave-voting Southwest England that they had lost in 2015 and had failed to recapture in 2017 and 2019.
We find that this year’s election results still reflect the Brexit cleavage – more so even than in 2017.
After the divisive struggle between the “Leave” and “Remain” camps that virtually paralysed government between 2016 and 2019 and culminated in Boris Johnson’s 2019 campaign slogan “Get Brexit Done”, we would have expected voting patterns in the 2019 general election to reflect the way people voted in the 2016 referendum. Specifically, we would expect Remain voters to have voted for left-leaning parties (Labour, Lib Dems, Greens, SNP and Plaid Cymu) that advocated either a second referendum or the cancellation of Brexit and Leave voters to have voted for right-leaning parties (Conservatives and the Brexit Party) that advocated a “hard” Brexit. Given the apparent low salience of Brexit in this July’s election, however, we would expect this effect to have since diminished. Instead we find that this year’s election results still reflect the Brexit cleavage – more so even than in 2017.
While precise data on how individual constituencies voted in the 2016 referendum are not available (because counting areas represented local authorities rather than constituencies), constituency-level estimates provided by Dr Chris Hanretty of the University of East Anglia are commonly used as a proxy. Dr Henretty provided estimates both for the former constituency boundaries and for the new (post-2023) constituencies. The diagrams below compare the differences over successive elections (2015, 2017, 2019 and 2024) between the combined vote of right-leaning parties (the Conservatives, UKIP, the Brexit Party and Reform UK) and the combined vote of left-leaning parties (Labour, the Liberal Democrats, the Greens, SNP and Plaid Cymru) with the projected 2016 Leave vote. The dots on the map represent each constituency and are coloured to represent the party that won the constituency. Only constituencies in England, Wales and Scotland are shown, due to the greater complexity of party positions in Northern Ireland.
Looking at the four diagrams, it would appear that the relationship between the partisan vote and the Brexit vote was significantly closer in 2019 and 2024 than in 2015 and 2017.
The extent to which the proportion of votes cast either for right-leaning or left-leaning parties reflects the way the constituency voted in the 2016 referendum vote can be seen in the extent to which the dots (i.e. constituencies) hug the diagonal black trend line in the four diagrams (above). If all constituencies are located along the line, we can say that voting behaviour is fully determined by voters’ positions on Brexit. On the other hand, if they are scattered randomly across the grid, we can say there is no relationship between the referendum vote and the general election vote. Looking at the four diagrams, it would appear that the relationship between the partisan vote and the Brexit vote was significantly closer in 2019 and 2024 than in 2015 and 2017 as the dots are more tightly concentrated around the trend line.
This intuition is confirmed if we calculate the R-squared parameter, which represents the proportion of the vote gap between left-leaning and right-leaning parties that can be explained by the Leave vote in the referendum. The R-squared values are 0.25, 0.26, 0.53 and 0.52 in 2015, 2017, 2019 and 2024 respectively. This means that while 25 per cent and 26 per cent respectively of variation in the vote gap in 2015 and 2017 can be explained by support for Leave in the constituency, this figure increases to 53 per cent in 2019 and slips back only to 52per cent in 2024. This shows that voting behaviour still strongly correlates with the referendum vote, even more so that in 2017 (the year after the referendum vote) and is little changed compared to “Brexit election” of 2019.
An opinion poll carried out by YouGov at the beginning of June showed that while 42 per cent of those intending to vote Conservative and 82 per cent of those intending to vote for Reform UK deemed immigration and asylum to be the most important issue, only 10 per cent of prospective Labour voters did so.
In 2024, as in 2019, we have two distinct blocs of voters that not only remain quite closely related to the old Brexit divide, but diverge significantly in the issues they deemed most important in the general election campaign. While Brexit itself was no longer seen as an important issue, asylum and immigration, issues around which the Brexit debate was framed, proved salient, but only to Conservative and Reform UK voters. An opinion poll carried out by YouGov at the beginning of June showed that while 42 per cent of those intending to vote Conservative and 82 per cent of those intending to vote for Reform UK deemed immigration and asylum to be the most important issue, only 10 per cent of prospective Labour voters did so, further emphasising the divergence of world view between the two groups of voters.
In an earlier article I wrote for this blog in 2019, I explained how these two groups of voters were of more or less equal strength. In the 2019 elections, the Conservatives won because Leave voters rallied around Boris Johnson’s Conservatives, while Remain voters were split between Labour and the Liberal Democrats. In 2024 however, it is the Leave camp that appears most split, this time between the Conservatives and Reform.
These trends should provide a stark warning for the Conservatives as they pick their next leader. A temptation will be to tack right and reunite the old Leave camp that Boris Johnson successfully relied upon in 2019.
While former Remain voters still flock to both Labour and the Liberal Democrats, tactical voting amongst both parties’ supporters seems to have maximised the seat count for both. Tellingly, the change in the vote won by the two groups of parties between 2019 and 2024 is not as great as overall electoral volatility would suggest; combined support for the Conservatives and the Brexit Party/Reform in England, Wales and Scotland fell from 46.8 per cent in 2019 to 39.1 per cent in 2024, while support for Labour, the Liberal Democrats, Greens, SNP and Plaid Cymru rose from 46.8 per cent to 52.0 per cent. However, a significant part of this change can be explained by the fact that the first group consists mainly of older voters, and a proportion of these will have died in the period between the two elections. Analysis by New Statesman data journalist Ben Walker estimated that by March 2024 11 per cent of those who had voted Conservative in 2019 had since died, compared to just 4 per cent of Labour voters.
These trends should provide a stark warning for the Conservatives as they pick their next leader. A temptation will be to tack right and reunite the old Leave camp that Boris Johnson successfully relied upon in 2019. If we assume there will be relatively little movement between the two groups of voters (as has been since 2019), it is likely that the combined strength of the right-wing bloc of parties will be lower still by the time of the next election (in 2028 or 2029) as a result of voter mortality.
As we can also see from the 2024 scatter plot (above), many seats that the Tories did manage to hold onto are outliers and have a higher Conservative/Reform vote than would be expected from their voting record in the 2016 referendum (see the blue dots above the black trend line). This is likely to be a legacy of traditional voting patterns in so-called “blue wall” constituencies. We saw in this election how voters in these constituencies began to wean themselves off that habit and, unless the Conservatives develop a new appeal to younger voters, they may continue to do so.
All articles posted on this blog give the views of the author(s), and not the position of LSE British Politics and Policy, nor of the London School of Economics and Political Science.
Image credit: John Gomez on Shutterstock.
