Most politicians – with the exceptions of the Lib Dems – have said that parliament should accept the results of the EU Referendum as the democratic will of the people.

This may be true for political or pragmatic reasons. Ethically, however, it’s far from obvious. If someone tells you Brexit is a moral necessity just because a vote has taken place, they are wrong. Obeying the vote requires a value judgement about the status of that vote, and the issue much more complex than simply asserting that a vote has taken place.

If we were to go around disputing the status of every vote, democracy would be impossible. Here I will present the case that the Brexit vote is uniquely precarious: direct democracy about an irreversible and highly important decision carried out in the context of asymmetric information. Specifically, polling data suggests many leave voters are expecting an outcome that not even the leave campaign itself think is possible.

There are good reasons to be wary of attempts to understand what voters ‘really’ wanted – analysis becomes a vessel for your own opinions. At the same time, we have to acknowledge that voters’ opinions can be shaped by the information they are receiving.

For these reasons, refusing to leave the EU would be an absolutely legitimate position for parliament.

In the national debate it seems to go almost unquestioned that simply going through a voting procedure automatically conveys unassailable democratic force to a decision. Not true: Russia, Zimbabwe, North Korea all have voting procedures, yet most people agree that they unsatisfactory in various ways. I’m not comparing the UK to those countries, but making the philosophical point that you stand in a booth and fill out a form and still not be ‘doing democracy’.

For a vote to carry democratic force – for it to convey the ‘will of the people’ – most people think you have to do more than just count pieces of marked paper. I complained about two criteria that I felt were lacking in the EU referendum before the vote took place – that the electorate be representative, and that voters should be well informed.

We apparently can’t agree on the demographics of EU Referendum voters, but we do know participation was unusually high, so let’s set the issue of representativeness on one side.

The electorate were not well informed, in fact they were actively misled about what leaving the EU would mean. This is the case in every election, here I will make the argument that the misinformation was both asymmetric and effective in changing voters’ views.

I’m also not claiming leave voters are stupid, or that they do whatever Rupert Murdoch says. I am not claiming that everyone who voted leave was misled. I am not claiming that voters would have voted remain with better access to information.

I am claiming that we do not know how voters would have behaved with better access to information, and that information in the EU referendum was unusually low quality.

This is a difficult empirical point to prove. We cannot observe how voters would have behaved in other circumstances. What we can do is build an empirical case that voters held beliefs that can reasonably be expected to influence voting behaviour, and that those beliefs are a result of systemic misinformation.

We can see from YouGov’s polls that many people believed that leaving the EU would make no difference to, or improve, the economy. In the last poll, which closed in the 19th June, 46% of respondents thought there would be no economic impact, while 9% thought they’d be better off. These views, unsurprisingly, correlate with the intention to vote leave. 18% of those intending to vote leave thought leaving would improve the economy, and 66% thought it would make no difference.

This is at stark variance with predictions. The Leave campaign’s economist Andrew Lilico own forecasts suggest that there would be a short term economic hit, but predicted that by 2030 the economy would have returned to normal. This prediction is more optimistic than almost any other, either from a private company, the Treasury or international organisations such as the OECD. If voters were aware that the most optimistic case was a short term recession, followed by a possible return to normal growth in 15 years time, rather than believing there would be no difference or an improvement, how would they have voted? We do not know.

This in turn bears on Leave’s promise to have extra money to spend on the NHS. A post-Brexit government can choose to spend more money on the NHS, they will not be doing so using ‘spare money’ created by Brexit – certainly not until 2030.

We are in the position of living in a future where Brexit now seems imminent, and the prediction of a short term slow down appear to be coming true, with Mark Carney confirming these effects both verbally and by providing £250Bn of tax payers money to support the economy.

In the same poll, 54% of respondents believed that Brexit would reduce immigration. Again, this correlates with intention to leave, with fully 85% of leave voters believing immigration would decrease. And again, this is at odds with the predictions of all sides. Leave’s economic model relies on immigration remaining roughly the same (Andrew Lilico again), and Leave campaigner Dan Hannan notably confirmed that immigration will remain broadly similar after Brexit. How would voters have behaved if they knew this? Again, we do not know.

I’m not claiming access to an objective reality about what will happen in the case of Brexit, instead I’m asserting that leave voters did not understand the position of the Leave campaign itself. Given that the Leave campaign is likely to have been over optimistic about what it can deliver, the reality of Brexit is likely to be even less satisfactory to leave voters.

We know that a typical leave voter thought that the economy would remain the same or improve while immigration would be reduced. But we do not know if these were factors that caused them to vote leave, or merely incidental. However, if we look at polls of issues that matter to voters, we see that immigration, the NHS, the EU and the economy are the top four issues. The average leave voter held unrealistic expectations about all of these, so it is reasonable to assume that some voters choose leave on the basis of these issues.

Where does these bad information come from? How can voters have come to believe a case for Brexit even more optimistic than the Leave campaign itself? We do see that newspaper coverage, which is traditionally on the right in the UK, is was strongly skewed to Brexit. Weighted for number for readers, newspaper articles about about 80% in favour of leave, even while the country as a whole is almost perfectly split. Meanwhile, the broadcast media are scrupulously balanced.

Article 50 has not yet been sent. The electorate now has a genuine opportunity to understand Brexit’s implications for the economy and immigration. If opinion polls show a significant shift in the light of this new information, that shift should be allowed to influence MP’s views; they should not feel bound by the referendum. The referendum did not convey an unassailable mandate based on the will of the people.

Edit: Reading Vernon Bogdanor I find my self slightly convinced by an idea similar to rule utilitarianism. Perhaps you can’t worry actual democracy in every vote, instead you have to set up the institution of voting and honour it regardless of the nuances of each referendum or election. Perhaps the damage to public trust is not worth the improvement in decision making.

4 thoughts on “The principled case for remaining despite EU Referendum

  1. Your argument appears to be based on the following assumptions:

    1. People vote rationally, that is to say, they vote based on available data and make rational decisions based on the information they acquire during the campaign.

    2. The available data for the referendum was flawed, in that the leave campaign misrepresented certain key facts (the infamous “350m” claim for example).

    3. Furthermore the media was biased towards one particular viewpoint and amounted to propaganda disseminating false information that led to a ‘leave’ outcome.

    4. After two weeks of negative economic headlines, we are in a position of superior information

    Therefore given that (2) the people were lied to, (3) the media was biased and (1) people act rationally, it is “principled” and even “ethical” to remain within the EU (your piece implies through the representative model of Parliamentary government, i.e. without a further vote).

    Kindly and wisely you mention the danger of allowing analysis to become a vehicle for your own opinions, so before I attempt to refute the argument above, I should state that –

    My desire to refute your argument most likely stems from my own moral position that, save for actual demonstrable voter fraud or other behaviour that breaks the written rules of the contest, the very fabric of a democratic society can only be maintained by respecting _every_ vote without exception.

    Once you overrule the outcome of even one vote, you set a precedent that enables democracy to be overridden at will. The logical outcome of this is either (eventually) a dictatorship or civil unrest. There can be no “do-overs” in a civil, democratic society.

    Yes, I know the referendum was de jure advisory, I believe it is however de facto necessary to carry out the result in order to maintain the fabric of democracy, for reasons stated above.

    This allowance for personal bias having been made, I’ll now attempt to explain why the arguments that lead to your conclusion are false.

    (1) On rational decision making


    People do not always vote rationally. Often, they vote emotionally. Brexit was as much a vote for a ‘feeling’ as it was a set of facts.

    Question – when people voted for Barack Obama in 2008, were they voting for a man, or voting for a feeling? (“hope”).

    Telling people ‘the economy is way better with Britain in the EU’ has no effect if people feel as if these benefits do not reach them, if people feel angry, dispossessed, etc.

    Furthermore while us metropolitan liberal elite types may find such notions distasteful, people still ‘feel’ patriotic, etc. We don’t have the right to discount emotional decision making as part of the process.

    We aren’t robots – we fall in and out of love. And it seems a lot of people have fallen out of love with Europe.


    When people do vote rationally, their rationale may seem inexplicable to us, as we are are unable to perfectly understand their choices – our own information is at variance to theirs.

    You might argue that if you told a person the consequences of eating a diet of Mars Bars, the fat content, the damage done, etc, that they would stop eating Mars Bars. The fact is that people continue to eat Mars Bars, despite the health risk. That’s because they like chocolate, it makes them feel good, etc. They prefer this pleasure to the pain of the long-term health risks.

    Your own argument based on economics is similar. You may argue that Brexit is enormously economically damaging (it has been so far, although the available data set is only two weeks old). However that discounts the fact that other people may consider this of secondary importance to other factors – being free from Brussels, having control of borders, lowering immigration etc.

    People have chosen an outcome that is not the best economically as they have chosen other factors to be more important to their well being.

    I frequently choose a pint over the gym for the same reason.

    (2) On the available data


    Information available for decision making is always flawed or, to borrow a more popular saying “hindsight is always 20/20”.

    This is not an acceptable reason for a do-over in a democracy, any more than it is in a game of cards – “If I knew he had aces, I would have folded”.


    Both sides lied – I won’t go into the specifics here suffice to say the lies on all sides are well documented. Perhaps Leave lied more or were more egregious. I don’t know. The 350m claim certainly seems so.

    But “the politicians lied” isn’t an argument to overrule the result.

    I would argue that most people are aware politicians lie / fail to deliver on their promises often and that each subsequent vote is an opportunity to hold the promises of the previous one to account.

    In these unique circumstances people were holding the government to account on a wide range of things, but specifically it seems likely that:

    – voters were holding the government to account for their pledges in 2010 and 2015 to reduce immigration

    – older voters were holding the establishment to account for promises made in the 1975 referendum


    People do not always vote based on the hard stats available to them, but rather speculate on future possibilities, extrapolated from that evidence. One could reasonably extrapolate, off the top of my head:

    – Immigration will continue to increase
    – The EU will continue towards further “ever closer union”
    – The single currency, Schengen, even possibly the EU as it stands, is a project that will eventually fail, leading to a systemic collapse it is better to thoroughly disentangle ourselves from now.


    One of the nubs of your argument is that people would choose differently if they had access to better information:

    You might argue that if the electorate only knew that immigrants accounted for 5% of the population, not 15% as is widely believed, the result would be different.

    But it is also widely known that net migration stands at 330,000 in 2015, the second highest figure on record.

    A leaver would say: I don’t care whether it’s 5% or 15%, I think 330,000 a year is unsustainable.

    You might argue that GDP has never been higher, employment has never been lower, etc. You might point to all these stats.

    These stats have very little relevance to someone who has not been a net beneficiary of globalisation, who is on a zero hour contract or in part time work, etc.

    I remember reading one very good interview with a leaver who was told by a visiting politician that it was ‘good’ he was back in work – he was a tax driver who had been a steel worker who was now earning 1/4 of what he used to earn.

    The stats do not always tell the full story.

    (3) Media bias is not a good reason to overrule an election

    The argument of media bias is so old now as to be a well-worn trope, while it’s true that most of the papers supported leave, I would also point your attention to the fact that

    – The Sun didn’t come out for Leave until the first polls came out demonstrating a decisive swing towards leave. It followed its readership rather than led it.

    – The government also attempted to stack the odds in their favour, for example with the infamous £9m HM Govt leaflet and using the resources of the civil service to make the case for remain prior to the purdah period.

    It’s always biased. You would hope the electorate realise this by now and adjust their filter accordingly.

    If The Sun told you to jump off a bridge, would you do it?

    (4) The economic situation post-referendum has changed

    The data set from which you are drawing your conclusions is too short.

    For example, if you look at the events of the last two weeks, of course you would say Brexit has been enormously harmful to the economy.

    In context, if you looked at Britain’s departure from the ERM on Black Wednesday two weeks later, of course you would say it was a disaster.

    The politicians are in the same position as us here: your argument seems to be ‘the last two weeks have been disastrous, democracy must be overruled’ – nobody knows what the position will be like in six weeks, six months, six years.

    Furthermore your argument assumes that economic considerations are the only factor here – questions of sovereignty, immigration etc that leavers may have voted on have not changed at all, let alone substantially, in the last two weeks.

    In conclusion:

    I’ll restate that my position comes from a moral position that democratic votes must always be upheld, even when we face negative consequences as a result of them.

    (A side note: As a comparatively wealthy, property-owning Londoner in working in a very international sector, Brexit has undeniably harmed my financial and career prospects considerably already, in fact it’s an unmitigated disaster thus far – but I don’t consider my personal situation a good enough reason to campaign for a do-over)

    I think if there is an argument to be made, it’s that


    52% is not a significant enough majority to make a binary decision, in a _liberal_ democracy such as ours, a compromise must be reached that respects the will of the whole electorate, rather than a simple tyranny of the majority, or else democracy is just two wolves and a sheep deciding what to have for dinner.

    For me that compromise is probably EFTA/EEA in the short term, which is probably where we’re headed. It won’t placate the frothing “leavers” but it is an accurate reflection of the wider polis.


    As the situation changes, and what ‘leave’ actually means becomes clear, there is a case for a second referendum on those terms – ‘leave the EU entirely’, ‘accept EEA/EFTA’, ‘don’t trigger article 50’. Let the people decide. I would also suggest deciding such a vote using AV.

    I suspect given Comres polling showing 4% of ‘remainers’ are unhappy with the result of the referendum compared to just 3% of ‘leavers’, you would either get a ‘leave entirely’ or ‘accept EEA/EFTA’ vote in the event of such a referendum.

    But there is no case to be made for overruling the results of the referendum. To do so makes a series of false assumptions as I hope I have demonstrated. And democracy is such an important social good we cannot allow our political class to overrule it.

    I’ll conclude with questions rather than a statement.

    Let’s assume parliament overrules the result.
    – Will that be the end of it?
    – How likely are we to see immediate civil unrest?
    – What conclusions would you draw, as a voter, about the political class and
    – How would you then respond at the ballot box in 2020? (Hint: Hello Prime Minister Farage).

    Respecting the outcome as it stands is the least worst option available to us.

  2. Thanks for your detailed reply, I agree with a lot of what you say.

    I had a paragraph about media bias, which is explanatory, but not a core part of my argument. Same for the paragraph about economic predictions coming true. So your points 3 & 4 are red herrings. If I wrote it again I’d omit those paragraphs, they just cloud things.

    On your point 1, I’m not claiming rationality for voters. If I thought they were fully rational I’d be assuming they were responding perfectly to all available information, and optimising everything they cared about. Instead, I assume a weaker condition: that many voters care about immigration, the NHS, the EU and the economy. This is in line with data from the Ipsos Mori poll linked.

    Do voters also vote on a ‘feeling’? Sure. I’m just saying the above listed issues were probably big factors for many people. I think this chimes with out intuition.

    I’m not saying the outcome would have gone the other way if voters had better understood Leave’s position, I’m saying we don’t know. As you say, patriotism (or an ’emotional part of the decision making process’) may have one the day regardless.

    On point 2, I’m very careful not to base anything on knowledge about the future. How can I claim to know more about the future than voters? Instead, I’m using facts about the present, in particular, what the Leave campaign itself promised. Which is: no substantial changes to immigration, and accepting there would be a hit to the economy.

    So the misinformation is not about what leave voters believed in comparison with my prognostications for the future. I’m claiming, using polls as evidence, that many leave voters did not understand what Leave itself claimed to offer. In particular, the typical leave voter thought the economy would remain unchanged or get better, while immigration would go down.

    I’m explicit that it’s an assumption that those misunderstandings influenced the votes. We can’t know, we can just say that those misunderstandings are highly relevant to the four issues people list as the most important to them.

    I do see the power of your point that ignoring even a single vote opens the door to wider breaches of democracy, as I note in my edit that may be the overriding principle here.

    I’m not sure about another vote on the actual nature of Brexit, it appeals to me a lot, but then I’m also not that into direct democracy, so I’m conflicted. I believe that if the vote were offered in a recession that people attributed to Brexit it might well be remain.

    I certainly agree that single market membership is important.

    I wouldn’t claim you can have a do-over of the vote on the basis of media bias, but it is a persistent problem, and the idea that voters can just filter it out is over optimistic. We can never get rid of it, but should newspaper ownership be so concentrated? Probably not.

    I did intend this to be a theoretical discussion. Pragmatically, at the moment, it certainly wouldn’t be desirable.

  3. It’s certainly an interesting thought experiment, taking away all the other factors and simply posing the question “if it could be proven that the majority of leave voters thought they were voting for lower immigration and a stable economy, things that now seem unlikely (due to falling economy + likelihood of EEA)” is the referendum invalid?

    I’d still argue no, for the same reason that the 2010/2015 elections aren’t invalid because the Tories broke their manifesto pledge to reduce immigration. People all too often vote for one thing and end up with another – that’s politics! The flip side of the coin is that most people who voted remain probably thought they were voting for things to remain the same – if we ended up with “more Europe”, such as a European Army or tax code, people would have been similarly misled.

    You get your chance to punish the people who lied to you at the ballot box next time. And punished they will be in 2020.

    And yes, I know leaving the EU is forever, but I think there’s a very strong case to suggest that turnout was high among the older generation and they were so pro-leave precisely because they remember 1975 and feel this is their first time to give their verdict on the promises made in that campaign.

    Perhaps we’ll be old men voting on our re-entry in 2057… In the meantime, democracy will muddle through all this somehow (I hope!).

    I remain completely in favour of a second referendum on the terms of our exit, with “don’t do it” as an option. But I’m an arch democrat and think all this stuff should be approved by the people.

    I suspect I’m in a minority in wanting another referendum though… I doubt the politicians will ever let us have one again!

  4. Political parties always fail to deliver all their manifesto promises. That might be a good thing – the point of representative democracy is for debate to inform decisions, so maybe it’s for the better when parliament tempers a manifesto promise.

    I’m not saying the referendum is unique because a promise will be broken. I’m saying the opposite. I’m saying leave voters were voting for something no one was promising.

    Second referendum is difficult because there is no ‘off ramp’ after Article 50 has been initiated, but before it is initiated there is no negotiation, so we wouldn’t know what options to put on the ballot. OTOH, is the EU really going to say no if the UK voted to remain half way through? Probably not.

    My favourite explanation for old people voting Leave is because they are voting to go back to the time when they were young again. I don’t know how many have waited 40 years to stick it to a bunch of politicians who are mostly dead.


Leave a reply

<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>