The logistics of a second referendum

Support for a second Brexit referendum appears to have risen of late. According to an opinion poll released yesterday supporters of a second referendum outnumber those who oppose it by a margin of 47% to 36%. Many politicians also support this option. As I have pointed out previously, I am by no means persuaded of the merits of a second referendum because it will do nothing to heal the divisions caused by the 2016 plebiscite. In addition it will be used by Brexit supporters as yet another example of how the political establishment conspires to foil the will of the people. Nonetheless, if parliament cannot resolve the current impasse and there is sufficient public support for this option, there is no reason in principle why it should categorically be ruled out.

But the logistics of the second referendum are such that it will be far more difficult to implement than the headlines would have us believe. The Constitution Unit at University College London set out in great detail many of the issues involved, and it is worth looking at this paper for those interested in the detail. With regard to the mechanics of the referendum, the first issue is how quickly it could be done and there are there four stages which will determine the front to back timeline.

In the first stage, parliament must pass primary legislation to provide the legal basis for a referendum covering issues such as the question on the ballot paper, conduct rules for the poll and the date on which the referendum will be held. There is no minimum time for this process but it took around 7 months for the first EU referendum to conclude. The analysts at the Constitution Unit reckon that on the basis of past experience, it may be possible to reduce this period to as little as 11 weeks but this would only be possible if there is cross-party consensus. In the current environment, this is far from the case.

The second stage of the process is question testing, which normally takes 12 weeks but the UCL team suggest that it could be done in 8 if needed. Stage three is preparation for the poll, allowing sufficient time to prepare for the administration of the poll and regulation of referendum campaigners. The Electoral Commission recommends a six month period but on the assumption that the same rules apply as in 2016, the UCL team reckon this could be kept to 1 week. The final stage covers the campaign period itself which must last for a minimum of 10 weeks (no exceptions). The UCL academics reckon that a minimum of 22 weeks are required if everything goes according to plan and if some of the processes are run concurrently.

On that basis, if parliament can agree on the need for a second referendum between now and next Thursday, they just might be able to get it done by 27 June – the last Thursday of the month. If we follow a sequential process, adding up the minimum time periods gives us a total of 30 weeks which takes us into the height of the holiday season in mid-August. The front-to-back timeline for the 2016 referendum was close to a year. In my view, the odds against getting a referendum done in 2019 are quite high.

Obviously, there is no chance of getting any of this done before 29 March. Thus if the UK does want to hold a second referendum it will have to ask for an extension of Article 50. It is generally supposed that the EU will grant the UK an additional three months but this would only take us to end-June, which is not enough time to hold a referendum and deal with the aftermath. An extension beyond this date becomes constitutionally complicated. The European parliamentary elections are held in May and the new parliament will convene on 2 July. However, the UK has not planned to send any MEPs to Brussels on the basis that it will have left the EU by the time the new parliament sits. If Article 50 is extended beyond June the UK will be in breach of its EU obligations if it does not send MEPs to Strasbourg to sit in parliament. A three month extension thus runs the risk of merely postponing the cliff edge until the summer.

However, it transpires that the Electoral Commission has set aside £829,000 for "activities relating to a European Parliamentary election in 2019" which was described as a “precautionary measure.” In other words, in case the UK needs to hold an election of MEPs on 23 May. Just imagine how that will play with the eurosceptics who will probably not bother to turn out to vote. Against that, it may enthuse Remain supporters to turn out in higher numbers than usual.  Whatever the merits of holding European elections, it is the only way that the UK can press for a longer Article 50 period that would allow the possibility of holding a second referendum.

Whilst it is theoretically possible to do this, nothing that we have heard from the political establishment – even prominent supporters of the second referendum – suggests they are even contemplating this option. But if they are serious about the so-called “People’s Vote” they will have to do so. Parliamentary opposition to holding European Parliament elections in the UK will be enormous and I can’t imagine for a moment that the government will support the idea. But this could be one of the next battlegrounds in the Brexit saga, the duration of which will surpass that of the Second World War in 12 days’ time.