Brexit Party doesn’t mean Brexit: 2 graphs that tell the real story
Much noise is being made about the Brexit Party’s rapid ‘success’ in European election polling. Both the pro- and anti-Brexit media are throwing around alarmist language like ‘terrifying’, ‘surge’, and ‘showdown’. All of their commentary is fundamentally misleading.
The reporting should really go something like this:
There is nothing surprising about the Brexit Party gaining 35% of the vote; this could easily have been predicted, at the very moment the party was established. In fact, if they achieve a figure of 35%, this would not be a demonstration of strength, but would show that Brexit has less support now than it did at the time of the referendum. And the evidence is not hard to find.
Here is the real story
This graph shows UK public opinion on leaving the EU since 2012. Of course, there are all kinds problems with the reliability of opinion polls, as we’ve discovered in recent years. Still, especially over a long period like this, they are measuring something; even if the absolute numbers are inaccurate, the trend can tell us something interesting. Moreover, if you look at the data around the time of the referendum in June 2016, it shows Leave and Remain as very close–a somewhat reassuring level of accuracy.
Why does this graph matter?
Well, first of all, it tells us that the British public is pretty evenly divided on the EU. For the past 5 years, about 40% of us have wanted to leave, and about 40% of us have wanted to stay. So the odds for one or the other to win a referendum would have been very close for a long time.
Another thing we can learn from the graph is that after a period of Leave holding steady at 50% and Remain at 30% (we don’t know how long this phase lasted before 2012), support for either side has fluctuated, with each overtaking the other pretty frequently.
These two observations–that opinion is roughly evenly split, and that it has fluctuated–suggest the following:
The result of the Brexit Referendum was determined primarily by the point in time at which the vote took place.
In other words, on 23 June 2016, Leave happened to be a bit stronger than Remain. A month earlier, or a month later, the opposite might have been true.
This feels like a pretty bold assertion. After all, these fluctuations in opinion might not have been arbitrary. It’s possible that they were influenced by the opposing sides’ campaigns, and by people informing themselves of the issues. It’s possible that at the last minute before the Referendum, the status of the debate, combined with current events, influenced a number of individuals to decide that Leave really was the best thing for Britain, pushing Leave over the top.
It would be interesting to test this position by analysing every fluctuation on the graph in terms of what was happening at the time, but most of the changes probably fall within the statistical margin of error (usually around 3%). This means that they may not be meaningful enough to indicate anything.
However, there are two key points when opinion switches dramatically.
The first comes roughly in February 2013; the second around May 2015. Both times, there is apparently a big jump in Remain sentiment and a big drop in support for Leave.
What happened at those points? On 23 January 2013, David Cameron announced that he would make holding an in-out Referendum part of the Conservative manifesto in 2015. And in May 2015, the Conservatives won a majority, making the Referendum inevitable.
This suggests a second interesting hypothesis:
When faced with the real – not theoretical – possibility of leaving the EU, the reaction of the British public was to shift to the position of Remain.
In fact, Leave never recovered from that initial drop in February 2013. So the question becomes, why was the same reaction not manifested at the time of the Referendum itself? Answering this properly requires academic investigation, but we can speculate about the scare-mongering, over-promising and obfuscation on both sides that accelerated after May 2015: confusion is said to be a factor that can influence referendum outcomes. And the notorious post-Referendum spike in online searches for “What is the EU” speaks of a certain amount of buyer’s remorse.
So would another Referendum make everything better?
Increasingly the case is being made that another Referendum needs to be held, based on the allegedly misleading nature of the Leave campaign, the political chaos that has ensued since 2016, and the unsatisfactory nature of Theresa May’s deal-making. However, in order to be worthwhile, the next vote would have to be a strongly decisive one.
Dominic Cummings, widely perceived as the orchestrator of the Leave campaign, has seemed confident that this would be the case:
“If there is another referendum, Vote Leave 2 will be much much worse for your side than VL1 was. VL2 will win by more than VL1…”
But as we saw above, if the polling tells us anything, it’s that there is a durable trend for the public to be evenly divided on this issue. This trend seems to be continuing since the referendum, showing a similar division when people are asked how they would vote in a second referendum. Again, both sides have been dancing around that 40% line, flipping and crossing at will. There’s been no immense change in favour of Leave to support Dominic’s confidence. If anything, Remain looks to be in the better position; and a 35% showing by the Brexit Party would be a decline from the level support for Leave earlier this year.
Maybe Dominic is counting on a bump from the Don’t Know/Undecideds (DKUs). However, there doesn’t seem to be a significant correlation here. Sometimes when the number of DKUs decreases, Leave and Remain numbers both increase together; sometimes Leave increases; sometimes Remain. There does seem to be a tendency for Leavers to have bigger drops and increases in tandem with the rise and fall in DKUs, suggesting that there is a greater number of marginal Leavers than marginal Remainers – but this may be a glib interpretation on my part. In any case, it’s not sufficient for a massive bump.
But how does Cummings buttress his argument? Let’s hear the rest of it:
“…At a minimum VL2 will win the referendum and destroy the strategic foundations of both main parties. The Tories will be destroyed and maybe Labour too. The rotten civil service system will be replaced and the performance of government will be transformed for the better. Investment in basic science research will flow. Long-term funding for the NHS guaranteed by law. MORE high skilled immigrants, FEWER low-skilled. An agenda that could not be described as Left or Right. The public will love it. Insiders will hate it but they will have slit their own throats and have no moral credibility. Few careers will survive.”
As it turns out, there’s not much more to the argument at all; these are the words of an emotional prophet, not a scientific forecaster. Cummings seems to have a powerful desire to convince someone – himself and/or the British public – that his personal dream of ‘draining the swamp’ will become a reality.
Even if an outright win by one side or the other was likely, however, there would be serious question marks over the legitimacy of such a vote.
What about democracy?
Leavers’ argument of last resort in support of Brexit is usually “But democracy! Respect the will of the people!” In making this argument, they are being either intentionally disingenuous – in service to their own personal ends – or remarkably naive.
Our review of the polling data leads us to one obvious sense in which referenda are undemocratic: they are a snapshot of an isolated point in time. We would not consider it democratic to vote in one government for the rest of eternity; that’s why elections are held every few years, because circumstances change, and people have the right to hold their elected leaders to account.
But what of the fact that a political party was elected that had made a promise to organise a Referendum? Does this election serve as a kind of permission granted by the people to hold the Brexit vote? Well, a Conservative government was elected in 2010 with a promise to repatriate powers from the EU, and it didn’t do it. Indeed, some Leavers point to this as a scandalous betrayal on the part of the Tories:
A 2nd one was the incoming Conservative govt of 2010 refusing to seek to repatriate any powers from the EU despite having run for four General Elections in a row on a promise to do so. https://t.co/wyIftwnnx3
— Andrew Lilico (@andrew_lilico) April 1, 2019
The argument doesn’t hold for the simple reason that the Conservatives weren’t elected in 3 out of those 4 elections; should we assume that they lost 3 of the elections because they made the repatriation promise, or that they won the 2010 election because of it? It’s impossible to establish a correlation; there were many promises made. The fact that a government is elected on a particular platform doesn’t necessarily mean that the majority of the population supports one specific piece of that platform.
Majorities, communities, and self-determination
This question of majority support is another obvious reason why certain types of referenda can be undemocratic – for a start, any referendum that doesn’t require a supermajority. James Madison famously called referenda the ‘tyranny of the majority’, but in the case of Brexit it wasn’t even that; it was the tyranny of the minority.
Brexit also raises important questions about the nature and size of the community that should have the power to decide its own fate. The nation-state is ultimately an arbitrary community that exists for historical reasons, and because it is recognised by the current global order. What makes it a legitimate unit of measure when it comes to public opinion? Brexit was a powerful illustration of this conundrum, given that London, Scotland, and Northern Ireland voted to Remain-with all the ensuing problems and debates. Questions that are voted on in referenda should be questions whose scope can be decided by a unit of people to whom the answer can usefully apply. In other words, the fewer and more closely connected voters are, the more likely it is that the outcome of a referendum will correspond to the interests of the community; it’s been found that direct democracy works best at the micro level.
This raises the larger philosophical issue of whether it’s ever right to put certain types of questions, which apply to a large polity – such as the sovereignty of a nation-state – to a public vote. Resolving this is beyond the scope of a single article, but it’s very important to consider, taking into account mixed evaluations of existing models of direct democracy in the US and Switzerland.
As a start, we can try to envision conditions under which a referendum on a decision as significant as Brexit would have maximum legitimacy:
- If it follows a sustained period of verifiable change in public opinion, thus responding to a real social and political shift;
- If it follows wide public calls for a vote, such as petitions (as in the Swiss model; this also avoids politicians calling a referendum for self-serving reasons);
- If the implications of the outcome of the referendum have been thoroughly researched, debated, and explained in clear terms to the public (even better if communities have been able to host meetings where members of the public discuss and debate the issue, as deliberation improves political efficacy and decision-making);
- If there are clear rules around, and impartial monitoring of, the referendum campaigning process;
- If the vote is won by a supermajority;
- And if the result is approved by Parliament (a ‘double majority’ in the Swiss system).
Brexit: Least-worst outcomes
I’ve tried to make the case here that the Brexit referendum is very hard to justify as ‘democratic’ in the sense that its most ardent defenders want us to believe. In a parallel universe, reasonableness would prevail, Parliament would recognise that a Brexit deal is proving intractable because there’s no sensible way for Britain to disengage from a system it has taken part in building for more than 40 years, and the Brexit Party would redirect its energies to reforming the EU from within. But the impacts of the referendum have grown so chaotic and convoluted that there are really no good outcomes in this situation. One MEP candidate has suggested it might be 30 years before the UK recovers economically from this debacle – and she’s from the Brexit Party.
Right now, a least-worst solution could be an early general election. The Conservative Party is imploding, Labour is incapable of effective action, the political landscape is shifting, the public is eager to be heard. An election feels like the only option to bring a modicum of clarity and renewed vigour to Westminster. If the Brexit Party gains a majority, then so be it; Brexit would have to be brought to its most radical conclusion. Cummings’ fevered vision of the established parties cleaning house might even come true.
What feels more plausible, however, is that in a national election people would be more cautious about voting for an untested party than in the European elections, which, let’s face it, don’t have a lot riding on them. A Brexit Party-Conservative vote split could leave space for a LibDem-Labour coalition and a mandate for a ‘soft Brexit’ or another plebiscite (which the LibDems would make a condition of any coalition). But a second referendum, even in the unlikely event that it could be held under the ‘ideal’ circumstances described above, even if it yielded a much stronger majority decision, would not necessarily put things right. There’s a good chance it would sharpen the growing divisions and mutual resentments in British society.
This brings us to the central challenge emerging from Brexit, which ultimately has very little to do with Brexit, and more to do with what the referendum has revealed about British society. Whatever the final outcome, the main task will be to move past the politicking and pettiness as quickly as possible, extract wisdom from what took place, and focus on building the things that will really guarantee British success in the 21st century: a diverse culture, progressive education, and scientific innovation.
