Is there time for another referendum before the new Brexit deadline?

alan.jfif (1)The last few weeks have been a whirlwind of Brexit news. Campaigners for a no-deal outcome have made themselves heard—as have campaigners for a second vote. But the new Halloween Brexit deadline is just over six months away. This raises the question: is there time to hold another referendum before we leave? And would it be possible to conduct such a vote in a proper manner? Alan Renwick addresses the key questions and concludes that a properly conducted referendum is preferable to a speedy one. 

In a report published last autumn, my colleagues and I at the Unit calculated that it takes at least 22 weeks—roughly five months—to hold a referendum in the UK. That allows 11 weeks for the necessary legislation to go through parliament and the Electoral Commission to assess the proposed question, one week to get ready, and ten weeks for the campaign. If parliament started this process today, a vote could be held on 26th September. So long as the wheels were set in motion by the European Parliament elections on 23rd May, a referendum could go ahead on 24th October; the last Thursday that gives time for the result to be declared before the deadline.

So the simple answer to the question posed above is, yes, there is time for a referendum by October.

But does pushing for a referendum at breakneck speed still make sense? Back when we were writing our report, the first question everyone asked was whether a vote could be held before Brexit day on 29th March. Once that timetable had become untenable, the question was whether the ballot could be organised by 23rd May or 30th June, so that the UK would not have to participate in the European Parliament elections. If a vote is being contemplated for September or October, that Rubicon will long have been crossed.

Crucially, EU leaders have signalled that the Halloween deadline will not be final if a decision-making process is ongoing by then. In other words, starting the referendum process could itself provide Britain with more time to deliberate. Businesses are desperate for some kind of resolution. But a well-run referendum would produce a more robust outcome. Taking a little extra time to ensure that would be worthwhile.

One fundamental issue concerns the options on the ballot paper. Campaigners for a referendum want a straight choice between the negotiated withdrawal agreement and the option of remaining in the EU. But many Brexiteers find the deal anathema. They would see such a vote as a stitch-up, delegitimising the result. A three-option referendum could alternatively be held, with a no-deal option alongside the other two. So long as a preferential voting system were used, that would be tenable. But most MPs see a ‘no deal’ Brexit as catastrophic. If voters chose it, would those MPs really be willing to acquiesce in all the preparations that would be necessary? If they cannot contemplate doing that, they should not put it before voters. There are no easy answers as to what the right approach is here—any referendum is fraught with dangers. It would be wise to spend time weighing the risks.

Then there is the question of how the campaign should be conducted. There is general acceptance that the rules by which the 2016 referendum was run were not fit for purpose. The Commons Digital, Culture, Media, and Sport Committee has called for reforms to tame the wild west of digital campaigning. The government’s recent Online Harms white paper acknowledges misinformation as a threat to democracy, and proposes a range of measures, including greater transparency for online political advertising. It would be remarkable if parliament did not seek to address some of these issues in the legislation enabling the referendum. Again, it would make sense to ensure proposals were examined in depth, not rushed through.

Whatever the legal rules, journalists, academics, policy experts and others should consider how they could foster a more reasoned debate than happened last time. Broadcasters such as Robert Peston have accepted that programme makers did not do a good enough job. Beyond fact-checking, which the BBC and others are already giving greater prominence, broadcasters should be promoting thoughtful discussion rather than endless seesaw tussling between the two sides. In another report, published last month, I argued that we should learn from Oregon, where, before a referendum, a group of randomly selected citizens meets for four or five days, hearing from experts and deliberating on the issues among themselves before setting out what they think in a short statement. The BBC could build on this model to draw the voices of ordinary voters into the heart of the campaign. Once again, setting this up would take time.

So, yes, a referendum could be held by October if parliament really wanted it. But rushing such a vote no longer makes sense. If a referendum is to be held, time should be allowed to settle the question, strengthen the rules, and do all that is possible to engender considered discussion.

This is a revised version of an article that originally appeared in Prospect and is reposted with permission.

About the author

Alan Renwick is Deputy Director of the Constitution Unit. He is the co-author of Improving Discourse During Election and Referendum Campaigns and The Mechanics of a Further Referendum on Brexit. He also served as Research Director for the Independent Commission on Referendums.

What would be the rules for a second Brexit referendum?

jess_sargent.000alan_renwick.000download.001This week’s Labour Party conference leaves a further Brexit referendum firmly on the political agenda. In the sixth of a series of posts on the mechanics of such a vote, Jess Sargeant, Alan Renwick, and Meg Russell examine what rules and regulations should govern the referendum process, arguing that important changes are needed to facilitate a fair and transparent campaign.

If  a further referendum on Brexit is held, the rules governing how it is conducted would be of utmost importance. The UK’s standing legislation on referendums – the Political Parties, Elections and Referendums Act (PPERA) 2000 – is both incomplete and in some respects out of date. As explained in a previous post, a new referendum would require fresh legislation. This therefore needs to fill in the gaps and update the rules to reflect the realities of modern campaigning. The natural starting point would be the legislation that paved the way for the 2016 referendum – the European Union Referendum Act 2015. But even that has deficiencies. This post examines key points that new referendum legislation would need to address. It also considers non-legislative changes that could improve the referendum campaign.

The franchise: who should be able to vote in a further referendum?

The franchise for referendums in the UK is not specified in PPERA, so would need to be defined in the legislation for a further Brexit referendum. The 2016 referendum franchise included all those eligible to vote in UK parliamentary elections, plus members of the House of Lords and EU citizens resident in Gibraltar. Some proponents of a second referendum argue this should be extended to 16- and 17-year-olds and EU citizens resident in the UK.

There are good arguments for extending the franchise, and precedent for doing so: 16- and 17-year-olds and EU citizens resident in Scotland could vote in the 2014 Scottish independence referendum. But – despite attempts to change this in parliament – the 2016 EU referendum legislation did not extend the right to vote to these groups, and consistency matters. If it appeared that the result of the 2016 referendum had been overturned because the franchise had been changed, many Leave supporters would view this outcome as illegitimate. As such, the franchise for any further referendum should be the same as for the 2016 vote.

How might referendum regulation be improved?

The referendum regulations in PPERA have not been substantively amended since they were introduced 2000. Since then, five referendums have been held, and the nature of communication and campaigning has changed significantly. Continue reading