The leave campaign’s surprise win in the EU referendum has led many people to look for explanations for its success and how it managed to perform so much better than UKIP, the party that pushed hardest for Brexit. At first glance, it may seem obvious that leave brought in large numbers of new supporters. After all, 52% is far higher than the the 12.6% UKIP managed in the 2015 General Election or even the 27% they managed to get in the 2014 European elections. On the other hand, aggregate data from the referendum show a very strong relationship between UKIP’s 2015 share and the results (at the counting unit level), as the following figure shows.
EU Referendum Counting units by 2015 UKIP vote share and % vote for Leave
The answer to reconciling these points is that the UKIP vote is extremely volatile at the individual level: from election to election and poll to poll. Lots of people stop or start supporting UKIP all the time. This means that the proportion of people who have ever supported UKIP is much higher than their share at any given moment. Looking at vote intentions from the start of the British Election Study Internet Panel in February 2014 to the general election on May 2015, UKIP only held onto 51.7% of their February supporters. However, they still performed reasonably well because 41.7% of their 2015 voters came to the party during that same time period as well.
So just how much of the leave share can be attributed to the UKIP curious? To answer this, we pooled seven waves of the British Election Study between February 2014 and May 2016, and looked at all of the vote intention and recall questions we asked for: general elections, European elections, local elections, devolved elections (including the London Mayoral race), and Police and Crime Commissioner Elections. Taking all of these together, we found that in this period 35% of respondents chose UKIP as their choice at least once. We can expand the pool of UKIP curious even further by using the propensity to vote questions that ask respondents “How likely is it that you would ever vote for each of the following parties?” with responses given on a 0 to 10 scale. We include anyone who gave a score of 6 or higher to UKIP in wave 7. This brings the UKIP curious percentage up to 38.7%.
The results are shown in the following figure, which shows the percentage of each side’s vote that is comprised of those who were UKIP curious since early 2014 and those who never showed any support for UKIP. On this measure, 67.1% of the leave vote consists of voters who have at least dabbled with UKIP. While there are some leave voters who never expressed an interest in the party, they are a minority of leavers.
These results show that when trying to understand the leave vote, we should be looking at the same things that make people support UKIP. While the Leave side vastly outperformed UKIP, it still mostly drew upon voters who have supported UKIP at some point. This fits with our previous research with showed that while UKIP disproportionately appeals to the dispossessed, it also manage to pull substantial support from less likely groups such as the middle classes and former Liberal Democrats. It also fits with our previous findings that anti-EU attitudes and anti-immigrant attitudes have become closely entwined and highly salient after the borders were opened to the EU accession countries. While these results may seem to suggest that UKIP has a huge potential base of support. It should be noted that the other major parties have even larger groups of voters who have shown at least some support for them during the last two years. Overall, these results serve to remind us that we live in an unprecedented era of volatile party support. The fate of UKIP (like almost everything else in British politics right now) is very uncertain, but if they do fail it won’t be because of a low ceiling of support.