Boris Johnson was responsible for shifting the electorate’s view of the Conservative Party in 2019? Wrong.
The standard narrative of the 2019 election is that Boris Johnson fundamentally altered the appeal of the Conservative Party, allowing him to reach parts of the electoral map previously alien to the Tories. This view has him boldly reaching out to traditional Labour strongholds and winning them over in a way no leader before him had been able to.
The data refutes this.
If you run a correlation analysis at a constituency level between the Conservative Party vote share and demographic and economic data it shows a different, but arguably more interesting story.
There has undoubtedly been a shift in the voting patterns for particular demographic groups since 2010, but the biggest change did not happen in 2019 – it happened in 2017 before Boris took charge of the Party. Broadly speaking, there was little difference between 2010 & 2015 with a big change in 2017 that continued into 2019. The current parliament reflects the conclusion of a movement that began well before the current leader took the helm.
The radar graph above shows the extent to which each constituency’s vote share correlates with a number of different variables. The output is number between -1 and 1. At 1 there is an absolutely positive correlation – both go up at the same time. At -1 there is an absolute negative correlation – as one goes up, the other goes down. At 0 there is no relationship between the two. I’ve coloured them in: green is positive, red is negative and orange is neutral.
What it shows for 2019 is what you might expect comparing it to 2010. The Conservative Party’s popularity in areas with higher education and wealth has dropped. It has grown in areas with older people, more white people. It grew in areas showing indicators of poverty – no passports, higher proportion of people with no or few qualifications, people not owning a passport. It also grew in areas with more people working in manufacturing. In addition it shows an increasing correlation between voting Leave and voting Conservative.
But the bulk of this change happened in 2017. Whatever his strengths and weaknesses, successes or failings Boris Johnson did not single handedly change the position of the Tory party in 2019 – he exploited a trend already underway.
This analysis pulls its data from several sources particularly http://data.parliament.uk for the election results, https://www.nomisweb.co.uk for the demographic data and https://commonslibrary.parliament.uk for estimates of the constituency level Euro Referendum results.
For simplicity and to remove the complexity added by the Nationalist parties, I’ve only used the English constituencies. There are too many alternative drivers of vote share – particularly in Northern Ireland.
The correlation analysis (and the bulk of the data sourcing) is written in R and the radar diagram produced in Excel. I have pulled out those variables showing the most change between 2010 and 2019 as well as a couple which haven’t moved as dramatically, but I believe demanded inclusion (specifically Age & % White).