The Case for a Second Brexit Referendum

The Case for a Second Referendum

I will start with basic uncontested facts:

  • The original referendum was not legally binding. The contrast with the 2011 Alternative Vote Referendum is the reference point here. The Bill for the latter referendum included explicit clauses binding the government to the results of the referendum, while the EU referendum bill had no such provisions. Neither did the EU referendum bill explicitly state that the referendum was only advisory.  However, in ruling on the question of whether the Government had sole authority to invoke Article 50, the High Court was explicit in finding that the results EU Referendum did not have the force of law in the absence of an explicit statement to this effect in the Bill.
  • The result of the referendum was very close, with 51.9% voting in favour of leaving the EU.
  • The official Vote Leave campaign broke electoral law during the Brexit referendum campaign by exceeding campaign spending limits – a fact that is now uncontested after Vote Leave withdrew its appeal against the original ruling. Another major Leave campaign organisation, Leave.EU, also broke electoral law, although this is still subject to ongoing appeals.
  • Both major political parties committed to carrying out the results of the referendum in the post-referendum 2017 General Election. Both the Conservatives and Labour featured such a commitment in their manifestoes.

In addition to the above, the following seem to me reasonable assertions, although they are probably less cut-and-dried than those facts laid out above:

  • Both the official Vote Leave campaign and other anti-EU campaign groups heavily featured claims about the EU that were untrue or misleading during the referendum campaign. This includes the notorious claim that the UK paid £350 million a week to the EU, which was the gross rather than net figure.  Likewise, the Armed Forces Minister Penny Mordaunt explicitly, but wrongly, stated that the UK would not be able to veto Turkey joining the EU if the UK remained. More generally, there has been significant contention after the referendum over how far claims made by Leave campaigners around funding for the NHS and the extent to which the UK could control immigration outside of the EU were promises or a ‘series of possibilities’.  It is of course hard to establish how far such claims were deliberate lies, but it certainly seems reasonable to assert that the Leave campaign heavily featured a series of misleading claims, whatever the intent behind them.
  • The extent of debate and discussion around Brexit after the Referendum has led significant numbers of people to change their mind. Most recently the long-established British Social Attitudes Survey found that 55% of its respondents would vote in favour of Remain were the referendum run again.  As John Curtice, one of the lead researchers on the survey notes, the margin of error in this survey is such that this is by no means definitive evidence that the majority of the population now oppose leave.  But it is certainly sufficient to raise reasonable doubts over the extent to which Leave remains the majority position.

Finally, I want to make another important assertion.  The original referendum was vague.  The exact wording of the referendum question was as follows: Should the United Kingdom remain a member of the European Union or leave the European Union?  In wording this question, several significant issues were left unaddressed, including the status of the UK within the European Economic Area (EEA, the ‘Single Market’), or within the European Union Customs Union (EUCU).  Both are viable options outside the EU: Norway is a member of the EEA but not the EUCU; Monaco is a member of the EUCU but not the EEA.

In this respect, it is worth noting a contrast between the Brexit referendum and the way in which referendums are typically used, particularly in constitutional change processes (which, de facto, Brexit is).  Typically, referendums are used (and in some countries constitutionally mandated) to ratify specific proposals that have already secured in principle backing of the legislative.  This was the case, for instance, in the Irish marriage equality referendum; and, indeed, is the case for every federal law passed in Switzerland.  Even when indicative referendums have been held in recent years, they have included more precision than in the EU referendum.  For instance, the wording for the referendum on the Alternative Vote may have appeared similarly vague: At present, the UK uses the “first past the post” system to elect MPs to the House of Commons. Should the “alternative vote” system be used instead?  However, the Parliamentary Voting System and Constituencies Act 2011 that legislated the (in this case legally-binding) referendum did include a precise definition and operationalisation of the Alternative Vote.  Likewise, when Australia held a referendum on becoming a republic in 1999, the full draft legislation was the subject of the referendum.  Even the earlier 1975 non-binding referendum on the UK’s continued membership of the European Community (as the EU was then known) had a precise reference point and confirmatory orientation, phrased as follows: The Government has announced the results of the renegotiation of the United Kingdom’s terms of membership of the European Community. Do you think the United Kingdom should stay in the European Community (the Common Market)?

Putting all of these aspects together, it seems to me there are two distinct but interlinked reasons to hold a second referendum:

  • The original referendum campaign was unfairly conducted, evidenced by the breaches of electoral law by the Leave campaign groups and by the prevalence of misleading claims (whether intentional or not) in the Leave campaign.
  • There are sufficient grounds for reasonable doubt that the majority continue to support leaving the EU, evidenced by the closeness of the original result, the unfairness of the campaign on which this slim majority was won, and the consistent findings of recent polls that suggest opinion has shifted.

Either of these reasons, it seems to, are sufficient to justify a second referendum; the two combined are even stronger justification. If the original referendum had been legally binding, the second claim might not be strong enough to warrant a new referendum.  But the referendum wasn’t legally binding.  If the result of the original referendum had been more decisively pro-Leave, the first claim might been deemed inconsequential to the overall result.  But the result was not overwhelming.  Hence, a second referendum seems entirely justifiable.

Arguments Against A Second Referendum

So far, I have outlined a simple but persuasive case for a second referendum.  Now, I look at the arguments that have been marshalled against a second referendum.  These arguments typically make the broad claim that a second referendum would be undemocratic, but there are important specific ways in which this claim is made.  For convenience, I will call this general position the ‘Anti-Democratic Second Referendum’ position, or AD2R.

Theresa May herself laid out this kind of argument in an September 2018 opinion piece in the Sunday Telegraph.  She resisted a second referendum thus: Neither will I give in to those who want to re-open the whole question with a second referendum. In the Summer of 2016, millions came out to have their say. In many cases for the first time in decades, they trusted that their vote would count; that after years of feeling ignored by politics, their voices would be heard. To ask the question all over again would be a gross betrayal of our democracy – and a betrayal of that trust.

This is a curious argument.  First, this argument is based on some slippery terminology and assumptions.  ‘Millions’ came out to vote, and in ‘many cases’ this was the first time they ‘trusted their vote would count’.  How many, exactly or even within a reasonable threshold (give or take 100,000, say), constitute these ‘many cases’?  We simply don’t know.  The turnout for the Brexit referendum was 72.21%. The turnout for the 2015 General Election prior to the referendum was 66.4%; the turnout of the 2017 General Election after the referendum was 68.8%.  So the turnout was certainly around five percentage points higher for the referendum than for contemporary general elections.  But we still don’t know why those extra people turned out for the referendum; it is pure posturing to claim that this was the first time ‘they trusted that their vote would count’.  Or is Theresa May asserting that people normally vote without such trust?  If so, this hardly helps her standing as the democratically elected political leader.

More fundamentally, however, it is really not clear – whether in Theresa May’s formulation or in any other formulation – why a second referendum would be a betrayal of trust.  If everyone who voted in 2016 voted exactly the same way in 2019, this would not be a ‘betrayal of trust’, it would simply be confirmation of that intention.  But if the government has reason to conclude that voters’ intentions may have changed, then surely it is more a betrayal of trust not to go back for a second referendum.

This is all the more the case given the vague wording of the referendum and its non-legally binding nature.  If the referendum had been legally binding, then it could plausibly be claimed a betrayal of trust to renege on it. If the referendum had been more specific in its proposals, then there might also be a stronger claim that the government was breaching trust by holding a second referendum, although this latter claim seems to me still quite weak.  But in the context of a vague and non-binding referendum, going back for a second referendum cannot reasonably be called a ‘breach of trust’.

Let me give a counterfactual and an analogy to illustrate this point.

First, counterfactually, imagine that the Government had agreed an exit arrangement with the EU that saw us leaving the formal institutions of the EU itself but committed us to continue paying our subs, continue free movement of labour, and continue membership of the common market and the customs union.  It seems entirely plausible to assert that many leave voters would see this as a betrayal of trust – the deal was nowhere near the ‘Leave’ that they had envisaged. But according to the AD2R position, there would be no betrayal of trust in passing this legislation, but it would be a betrayal of trust to double check whether this legislation met the expectations of Leave voters.  It might be claimed that this is a fanciful counterfactual, but I would argue that, in fact, this is precisely the position we see being articulated by the ERG and the majority of Brexiteers in parliament in respect to the actual deal – they support Leave but not on the terms that Theresa May has negotiated.  So for the AD2R position to have any merit in this respect, its proponents need to justify why Brexiteers in parliament have the right to wrest control of the legislative agenda from the Government and to continually reject a specific set of proposals, but that not only do the voting population at large not have this right, it would be anti-democratic to grant them this right.

By extension, it is worth noting the contrast between Government outrage when Speaker Bercow invoked parliamentary rules to prevent parliament being asked to vote on the same proposal repeatedly with the same Government’s outrage at the prospect of the voting population actually being granted precisely the same opportunity.

Secondly, consider a simple individual-level analogy.  Imagine you have gone to a bank to enquire about a mortgage.  The bank has outlined the terms of the mortgage, and you have said that you would like such a mortgage.  Imagine now that the bank insisted that you had agreed to take a mortgage with them, then showed you the fine print of the mortgage, but asserted that it would be a betrayal of your trust to ask you to confirm that you wanted the mortgage on those terms.  This would clearly be ludicrous.  Yet the AD2R position is making a precisely analogous claim.  Leaving aside the valid concerns about the conduct of the referendum noted above, the voting population have said to the government they would like to leave the EU. The government has negotiated terms for such a departure. It now seems entirely reasonable to ask for the opportunity to have a say on those terms.

A second dimension of the AD2R position explicitly links the claimed obligation to leave without a second referendum to the conduct and outcome of the 2017 referendum.  The official Government response to a Parliamentary petition for a second referendum held this position: in the 2017 General Election… over 80% of people also voted for parties committed to respecting the result of the referendum. In fact, both major parties stood for election on a stated policy to respect the decision of the people. The Government is clear that it is now its duty to implement the will expressed by voters in the referendum – respecting both the will of the British people, and the democratic process which delivered the referendum result.

Now it is true, as I noted at the outset, that both Labour and the Conservatives committed to carrying out Brexit in their election manifestoes.  But here’s the problem with this argument.  Technically, but importantly, voters in general elections don’t vote for a Government, they vote for a representative in Parliament. The largest party or coalition in Parliament then forms the Government.  As the continued rejection of May’s deal and the outcome of the Letwin amendment votes demonstrably show, this Parliament cannot agree with a majority on any way forward with Brexit.

From the AD2R perspective, then, it could be claimed that Parliament is indeed failing to deliver on the manifesto commitments of a large majority of its members.  But in fact, the majority of Members of Parliament did vote for Leave options during the Letwin amendment debate; they only differed on which option they preferred. Just by putting together the votes for two of the indicative leave options – the No Deal option and Labour’s Alternative Plan – gives 396 votes (61%) in favour of at least one of the options.  (There were 160 votes for No Deal and 237 for Labour’s plan, with only one MP – Dennis Skinner – voting for both.)  And this is with only two of the options aggregated.

So, Parliament is clearly strongly in favour of Leave, but is gridlocked on how to leave.  It is completely inconsistent, however, to maintain that this gridlock is somehow undemocratic, failing to respect the will of the referendum, or breaching manifesto promises.  As I have continually noted, the referendum was vague, so it is not a ‘betrayal of trust’ if parliament cannot agree how to implement it.  And while both major manifestos committed to Leave, they staked out very different visions of what that meant.  So the failure of a hung Parliament to find a majority position is not a betrayal of manifesto positions, it is largely representative of those different manifesto positions.

This is not to say that the current parliamentary impasse is desirable, but it is the outcome of a democratic process that elected a range of members from parties committed to different visions of our future with Europe.  Including, of course, several parties committed to continued EU membership.

From whichever position you approach the argument against a Second Referendum, then, its claims really only hold any force if you imagine the results of the first referendum to have been definitive, precise, and unambiguous.  Yet, as we have seen, it was none of these things.

What Should a Second Referendum Look Like?

Given the above, there are two clear conditions for a second referendum: it must be precise, and it must be legally-binding.  The problem is that there are a range of options in play.  Specifically, it seems to be that there are three precise options that could be legislated and one more vague option that needs to be specified for a referendum.  The three precise options are:

  1. Exit on the terms agreed between Theresa May’s Government and the European Union. This deal could easily be given the force of law through a referendum, because it has already been agreed to by the EU.  It would be the kind a ratification referendum described above.
  2. No Deal Exit. Again, this option is quite precise. It would require the Government to inform the EU that (a) there would be no further extension sought for Article 50; and (b) no further negotiations would happen on an exit deal.  No Deal would then happen by default.
  3. Revoke Article 50 and remain within the EU. Again, this is a quite precise option that could easily be legislated for, particularly since the European Court of Justice has explicitly ruled that the UK can revoke Article 50 unilaterally.

Option 4 is to commit to Leave but to continue negotiating a new deal with the EU.  This is much harder to frame precisely, but could be achieved along these lines:

  • Commit to extend Article 50 for a set period of time (say 1 or 2 years) with no possible further extension and no possibility of revocation;
  • Dissolve Parliament and call a General Election; and,
  • Authorise the new Government to negotiate a deal to leave the EU within this time frame without the need for parliamentary approval.

This option would hence create a specific mandate for the Government to negotiate a new deal, while providing a precise timeframe and outcome (No Deal Brexit) if it fails to do so.  The General Election is required because otherwise this option would simply result in the May Government signing its previously agreed deal, which is Option 1 not Option 4. Note that this option would also rule out a ‘backdoor Remain’ because the of non-extension and non-revocation of Article 50 clause.

With four options on the table, a staggered set of referendums might be logical, but it seems to me much more efficient and representative of precisely the dilemma we currently face to enact this through a preferential voting system.  Voters would be asked to rank these options from most preferred to least preferred.  Least popular options are then eliminated and votes redistributed until an absolute majority is achieved for one option.

Such a referendum, it seems to me, would answer concerns on both sides.  For those committed to Leave, it would provide the clarity that is currently lacking about the popular will over how to leave and in doing so break the parliamentary impasse.  For Remainers who believe that the results of the original vote were unfair and/or that the popular will has now changed such that the majority are in favour of Remaining, it would provide the opportunity to establish the veracity of this position.

Either way, I cannot see anything undemocratic about a Second Referendum and indeed it seems in fact the most democratic way to break the current political impasse.