Robert Hoskins looks at how a vote should be structured, learning the lessons of 2014 and 2016, in the event of a second independence referendum.

Since the morning after the result of the EU referendum which marked a discrepancy in voting behaviour between Scotland and England, the demands for another independence referendum have been ever increasing. The constant nationalist drumbeat for indyref2 has intensified dramatically since Lord Ashcroft’s poll gave Leave UK a 52% – 48% lead. This poll marked the first lead in favour of independence since March 2017.

In marked contrast to the clamour for a second indyref there has been absolutely no debate whatsoever on what the structure of the next referendum should be. Perhaps the reason for the absence of debate on this is because the nationalists assume that it will be an identical rerun of September 2014 consisting of a one vote event, based on a win requirement of 50% + 1, with no turnout threshold or confirmatory vote required. Indeed the nationalists’ referendum bill which is currently being fast tracked through Holyrood suggests just that. The Bill assumes that a repeat of the one-off 2014 referendum is a given and only includes provision to set the date, the referendum question (which the Electoral Commission wants to change from Yes/No to Leave/Remain) and the referendum period  by secondary legislation.  Surprisingly, there has been no challenge about the structure of the next referendum from other political parties, which would suggest that they too assume and are happy with it being a rerun of September 2014. 

Have parties actually rigorously scrutinised current academic thinking on what best practice should look like, in light of an impending No Deal Brexit, for the structure of the next referendum should there be one?  My guess is that they have not and as a result we are sleepwalking the nation into a second referendum which will be an exact replica of the first, thus depriving the public of an informed debate on what lessons if any have been learned from the Brexit debacle.

If parties have engaged with what constitutes current best practice in implementing referendums in the UK, they will have encountered the recommendations outlined by the Independent Commission On Referendums (ICORs). The Constitution Unit of University College London established the Commission post Brexit in 2017 to review the role of referendums in British democracy. The Commission comprises of 12 members who are made up of a variety of former cabinet level ministers, current and former parliamentarians including the Tory MP Dominic Grieve and the former chair of the Electoral Commission. The Commission is an advisory body only, with no power to force governments who are conducting a referendum to implement their recommendations. However, its recommendations on how British referendums are conducted are the nearest to a gold standard endorsement that is currently available in the UK and are therefore worthy of  discussion.

How does the 2014 referendum stand up with regards to what the ICORs now considers to be best practice post Brexit? Let’s break down the structure of the first independence referendum into its constituent parts – a majority of 50% + 1, no turnout threshold and a one off ballot with no confirmatory 2nd vote attached, and scrutinise each in turn.

Majority of 50% + 1

What minimal debate that there has been about the structure of the last referendum has tended to focus on the Edinburgh Agreement’s aspiration that the 2014 referendum should reflect ”a fair test and a decisive expression of the views of people in Scotland and a result which everyone will respect”. It’s a bit of a stretch to suggest that 50%+1 reflects a ”decisive expression”, and the SNP’s constant clamour for a 2nd referendum does not reflect a ”result which everyone will respect”. Therefore it is somewhat surprising to say the least to find the ICORs recommending maintaining the status quo. The rationale behind this decision is equally as baffling as it equates a one vote majority in a parliamentary election or passing legislation by a 1 vote majority as equivalent to a referendum won by one vote more than 50%.

Turnout threshold

Currently there are no turnout thresholds recommended for cessation referendums. The ICORs recommend against having a turnout threshold for several reasons. Most obviously, opponents of change would advocate a boycott of the ballot – something which would likely occur if the SNP advocated an unauthorised independence referendum. Turnout thresholds also require accurate electoral registers to produce legitimate results. Even small register inaccuracies could therefore compromise results.     

Confirmatory vote

Due to the extension of article 50 and the very real threat of a No Deal Brexit, the consequences of which were never debated in the run up to the EU referendum itself, all parties apart from the DUP and the Tories now want a confirmatory referendum to let the electorate have the final say on whether they wish to remain in the EU or leave.

The ICORs endorses the concept of a confirmatory vote in two scenarios. Firstly they recommend that:

”Any legislation enabling a pre-legislative referendum should set out a process to be followed in the event of a vote for change. If a government does not produce a detailed White Paper on the proposals for change, a second referendum would be triggered when the legislation or treaty implementing the result of the first referendum has passed through the relevant parliament or assembly.”

“In cases where a government does produce a White Paper detailing what form of change it expects to secure, the second referendum would be triggered only in the event that there is a ‘material adverse change’ in circumstances: that is, if the expectations set out in the government’s paper are not fulfilled. It would be for the parliament or assembly that called the referendum to determine whether such a ‘material adverse change’ had occurred.”

”The process to be followed should be specified in the legislation enabling the first referendum, so that the requirement for or possibility of a second referendum, and the reason for it, is clear to the electorate before the first vote takes place. The Commission’s recommendation hence applies to future processes of change requiring a referendum, and is not intended to apply retrospectively.”

(Report of the Commission on Referendums 2018 p203 recommendation 20)

Prof. Meg Russell, the Director of the Constitution Unit, has also stated that this statement should apply to any future Scottish referendum.

If we now assume that the next independence referendum applies this first statement, what would it look like in practice? It would mean that the preceding white paper would have to alert the electorate to the possibility of a confirmatory vote if a ”material adverse change” were to emerge from what was outlined in the white paper to what was delivered when the deal was signed off?  To enable this to occur the SNP would have to produce a detailed white paper presumably outlining the following: what currency they would use; the amount of UK debt they would take on, what interest payments on said debt would be, amount of foreign currency reserves required, start up costs of new state, size of public spending cut backs needed to account for £13.4 billion fiscal deficit and to compensate for loss of fiscal transfer, amount of Capital flight etc. A ”material adverse change” would kick in when the sign-off deal did not resemble promises made in the white paper as is almost certain to be the case in the above example.

But would an SNP government who called for the referendum run the risk of triggering a confirmatory vote against its own proposals even if the deal they got from Westminster did not resemble the deal they envisaged in their white paper, therefore putting independence itself at risk?  Of course they wouldn’t. A combination of extra parliamentary activities would immediately kick in to articulate the anger of the masses. Activities such as the signing of an online petition demanding a confirmatory vote and mass demonstrations, the size of which would easily dwarf all Under One Banner demonstrations put together.

The second example given by the ICORs is far more straightforward:

”Our recent reports have highlighted several key steps that could be taken. One point concerns the structure of the decision-making process. I have argued – building on the proposals of the Independent Commission and the Council of Europe – that, where a referendum must be held on a principle rather than a worked-out plan (as in the case of Brexit or Scottish independence), a process should be set out requiring a second vote on the final agreed deal. This would protect voters’ right to make an informed choice”

(Renwick A, Scotland’s plans for doing referendums better: an assessment – The Commission Unit 2019)

The above second definition would guarantee a confirmatory vote on the final agreed deal. This definition would likely cover any future independence referendum which omitted to give the level of detail set out in example one and offered broad brush stroke policy commitments instead.


ICORs provides much welcomed Gold Standard recommendations for the future conduct of the implementation of referendums. It would appear that the 2014 Referendum in its current format is no longer fit for purpose, as the Brexit experience has changed the constitutional make up of future referendums.  If Remain in the UK is to campaign for changes to be made to the 2014 Independence Referendum construct, those recommendations must be rooted in best evidence. To ignore best practice would leave Remain in the UK wide open to the accusation of trying to gerrymander the next referendum to make it impossible for Leave UK to win. That is why Remain in the UK must accept all ICORs recommendations as they will provide much needed legitimacy and gravitas to their case.

Of all the components of the 2014 referendum which the ICORs template has scrutinised, the Commission has provided a much welcomed endorsement for the concept of a confirmatory vote. Arguing for a confirmatory vote provides Remain in the UK with the best case and the best chance for successfully changing the ground rules for the next independence referendum.

It does this in the following ways: the public is now very familiar with the concept of a confirmatory vote as it has been widely debated in the context of Brexit. Repeated opinion polling suggests that there is a majority in favour of a Brexit Confirmatory vote; a precedent has been set as a Brexit confirmatory vote has already been endorsed by all political parties apart from the DUP and the Tories.  The 1st Minister herself has repeatedly called for a Brexit Confirmatory vote making it rather difficult for her to refuse calls for including one as part of the next Scottish Referendum. There is still an outside chance of a confirmatory Brexit vote actually occurring after October 31st which would establish a rock solid precedent.

By far the most important reason of all for Remain in the UK to lobby hard for a confirmatory vote is that it raises the bar of electoral scrutiny so high that there is next to no chance of a unicorn economic case outlined in a white paper being able to stand up to the intense examination of a referendum campaign and remain intact at sign off. After all the SNP’s Growth Commission which was given the brief of outlining the best economic case for an independent Scotland took 3 years to report back its findings and three days for a layman to dismantle them. A dismantling of such forensic filleting that its findings have been endorsed by a slew of economists. 

In conclusion, a confirmatory vote makes campaigning for a supermajority moot as it kills the concept of achieving independence by means of a referendum stone dead due to the Nationalists having an inherently weak economic case.