Hugh Wilson and Emma Macdonald, Professors of Marketing at Warwick Business School, dig deep into the survey data to explore voter shift and the argument for a second referendum on Europe in the UK.
Edited from an original article appearing in May, 2019. With kind acknowledgements to Warwick Business School.
Britain is paralysed by Brexit. Three years after the referendum that gave a narrow ‘Leave’ win, parliament remains in deadlock over what kind of Brexit it wants, and whether that deal should go back to the people in a second referendum.
Brexit’s effects are not just in the future, but in the present, too. Internationally, the UK is already much diminished. It has long aimed to punch above its weight, positioning itself as on the intersection of Europe and the US. Its energies now, though, are inward. This, in turn, sucks time, energy and confidence out of the whole EU bloc.
The public blame the stalemate on politicians considering their self-interest rather than that of the country. The Conservatives suffered badly in May’s local elections. The gains didn’t go to Labour, which is perceived as vacillating and mealy-mouthed on Brexit. Instead, they went to two parties offering unambiguous support for a second referendum: the LibDems and the Greens. For the first time, the environment is becoming significant in British voting, and the coordinated action the EU can offer is being seen as helpful from a green perspective. On the right, too, there is a re-alignment of voting around parties offering clarity on Brexit, with Nigel Farage’s Brexit party making spectacular gains in the EU elections. The UK’s MPs have to find something on which a majority can agree. Have voters already found a way forward?
Have the people changed their mind?
In many ways, voters are as split as parliament. But there is one way forward that does command a majority: remaining in the EU after all. In the last twelve months, a remarkably stable lead for Remain of around 8% has emerged. By contrast, leading up to the June 2016 vote the polls swung widely, indicating a voting population still uncertain of their views. The charts below plot the National Centre for Social Research’s “poll of polls” on referendum voting intention over a six-month period in each case.
The top chart shows the six months leading up to the 2016 referendum. The bottom chart shows the six-month period up to May 2019. A glance at these graphs shows a clear difference in the changeability of voters’ views. In just two months before the 2016 vote, opinion veered between a Remain lead of 10 per cent and a Leave lead of six per cent. Since November 2018, by contrast, the Remain lead has been rock-steady at six to eight per cent.
But do these polls reflect how people would actually vote? After all, humans exhibit many attitude-behaviour gaps – the bane of market researchers. We developed a statistical approach to this question in a recent academic paper analysing data from the UK’s 2010 General Election. We tracked a panel of floating voters over four weeks, and compared how they said they would vote with how they actually did. We found we could best estimate the actual vote by considering voter preferences over time, weighing the most recent preferences most heavily. It turns out you can’t just look at the latest polls: someone who flipped yesterday may flip back tomorrow.
Adapting this to the last four weeks of polling data suggests that a second referendum tomorrow would yield a Remain win of 54 per cent to 46 per cent. (The same approach applied to the 2016 polls correctly yields a narrow Leave win, unlike much commentary at the time.) The UK is still starkly divided, but it is no longer a statistical dead heat.
Substantial shifts on the economy and immigration
What’s changed? To start with, people are clearer on what they think about the EU. When surveying customers, good marketers consider not just attitude valence but also attitude strength – how strongly people hold their view. And if you ask people to vote, as happened in 2016, that itself is an intervention that forces people to make up their mind to an extent. Greater attitude strength now accounts for the far stabler polling pattern.
As for the increase in Remain preference, demographics are of course a factor: every year, around 455,000 remain-dominated young people reach voting age, while the 480,000 who die are skewed towards leave. But how about all the other voters getting older? Some indeed have switched their allegiance: only 88 per cent of Remainers would vote the same again. But even more Leave voters would switch, only 82 per cent saying they would repeat their vote.
Two reasons are likely. Firstly, many of those Leave voters now think very differently about the likely economic impact of Brexit. In May 2018, 42 per cent of them thought that leaving the EU would make the economy better, and only 16 per cent worse. By January 2019, that lead had evaporated (26 per cent ‘better’ and 27 per cent ‘worse’), according to Opinium polls.
Secondly, UK attitudes towards immigration have shown a dramatic reversal in the last few years. In 2011, 64 per cent of people thought that immigration had had a generally negative effect on the UK. By 2019, according to Ipsos MORI, that had reduced to 26 per cent. The inflection point where those thinking immigration was positive began to outnumber those regarding it as negative was in 2016, shortly after the Brexit vote. Put these factors together, and of the 17.4m who voted leave, only 13.5m are still alive and intend to repeat their vote.
Would a referendum be a done deal?
After three years of obstinately inconclusive data, then, we can now say that those who would prefer to remain in the EU clearly outnumber those who wish to leave. That might help explain the difficulties in parliament with signing off a Brexit deal of any sort: In June of this year, 67 per cent of MPs’ constituencies now have a Remain majority, according to Populus.
This does not, of course, guarantee that any second referendum in a few months would follow the current polls. To start with, one side might run a better campaign than the other – Leave being widely acknowledged to have run the better campaign in 2016.
Also, to vote one has to turn up – quite different from answering the phone to a pollster. Estimating turnout is currently more of an art than a science: polling companies put far more effort into establishing your preference than in estimating whether you will actually vote. In our General Election study, we found that turnout depends not just on your declared likelihood of voting but also on whether you voted last time (a habit effect), as well as whether you care about highly-charged topics such as climate change and terrorism. We don’t have enough data from the pollsters to repeat that for referendums.
A third variable in any referendum is the exact question. Ask voters to compare staying in the EU with a specific leave option, such as former PM Theresa May’s deal, and the Remain lead increases further. And finally, of course, there is the possibility of events prior to a vote swinging opinion.
Arguments about the legitimacy of another vote are, of course, another story. Some argue that any shift in opinion is irrelevant and that the 2016 poll decision is irreversible. Others say the essence of democracy is periodically checking in with the people. The longer the gap becomes since 2016, the more that argument gains support. A majority now favour some kind of public vote, though views differ on quite what that should look like.
What could gain a majority in parliament?
So will MPs align with the public’s preference for a referendum? The best guide is the last round of ‘indicative vote’ options MPs considered in March. Closest to being carried was the proposal that any deal should include a customs union, failing by only eight votes. The other near-miss was the proposal for a confirmatory vote on any deal, attracting the most votes, at 268, though with 295 against.
A customs union, then, might yet carry a cross-party majority. But leaving with a customs union would please very few among the UK’s polarised voters. Most want to stay in the EU; most of the remainder want to leave with no deal. For such a deal to progress without a public vote would leave Labour and the Conservatives owning this outcome in the mind of voters. It seems inevitable that they would therefore be hammered in the polls for many years to come – particularly as the process of leaving would inevitably be turbulent, whether one agrees with its desirability or not. With a public vote, however, these main parties would have a chance for the people to be seen as owning the result, as happened in 2016. It is strongly in their interest to offer one.
In a highly turbulent situation, one thing seems remarkably stable: the 48 per cent has become the 54 per cent. This will provide ammunition for those arguing for the legitimacy of a second referendum.
- View Prof. Hugh Wilson’s academic profile
- View Prof. Emma Macdonald’s academic profile
- Discover the learning opportunities at Warwick Business School
- Download Hugh Wilson’s article and many others in the Global Voice magazine special Europe focus.
Discover Responsible Business and Leadership. Download Global Voice, the Council’s quarterly eMagazine:
The Council on Business & Society Global Alliance is an international alliance between six of the world’s leading business schools and an organiser of Forums focusing on issues at the crossroads of business and society – The Council Community helps bring together business leaders, academics, policy-makers, students and journalists from around the world. Follow us on Twitter @_The_CoBS . Visit the Council’s website for a host of information, learning opportunities, and free downloads.