DEBUG: blog_post
Using the extension well - a six month time-table
13 Apr, 2019

So little time!

24 weeks – five and a half months - is the minimum time to hold a referendum. So Parliament needs to decide quickly what it wishes to do and the high level time-table for achieving it. London4Europe Committee member and former Home Office senior civil servant Michael Romberg sets out the skeleton of a time-table.

The article complements an earlier argument for what process to follow. A separate article shows how we would use a time-table if the EU were to give us a one-year extension.



The key text on how long a referendum would take is the UCL Constitution Unit report “The Mechanics of a Further Referendum on Brexit”.

Parliament needs to decide to do it. It would be possible for the indicative votes process to determine which options should be put to the electorate. MPs would however have to be generous and allow some options to go forward to the referendum that were less popular in the House if they had significant public support.

Parliament needs to pass a law. I suggest that in order to save time at the start and allow a more considered passage of the procedural law, there should be two laws. One would quickly set up the core elements of the process (citizens assemblies, two stage referendum) and the second with more time the rules for the referendum. Legislation could take about 11 weeks, but for a simpler core bill could pass more quickly. The more that controversial matters could be left to a second bill the easier it would be to have the first bill pass quickly. The more that controversial matters can be dropped from the second bill, the faster it can pass.

Opponents of a referendum would try to delay the process. It will be important to generate support from all quarters for the process. A two-stage referendum should reconcile Brexiters to the process because it will enable them to put their preferred Brexit to the people.

The Electoral Commission needs about 8-12 weeks to test questions for intelligibility and fairness. That can take place before Parliament has decided in legislation what the questions  should be because there are only so many reasonably likely possible options.

Preparation for the poll on the part of the Electoral Commission and campaigners could take about six months from the passage of the legislation – but could also be safely compressed to 12 weeks or so.

A ten week campaign period of which the first six weeks are to allow lead campaign bodies to be designated and four weeks for the actual campaign – though the Electoral Commission has recommended that the four week campaign period be extended to ten weeks.


Citizens’ Assemblies

There will not be time for a Citizens’ Assembly to assist Parliament in determining the questions in the first round of the referendum. So the rôle would be to inform the public debate on the questions chosen by Parliament.

Citizens’ Assemblies need time to do their work. The Irish abortion referendum Citizens’ Assembly met over five weekends. Brexit is for sure more complicated. A citizens’ assembly on Brexit would confront many people with new information for the first time. Twelve weeks would be better but there is not the time. I therefore suggest one should allow two months, ie 8 weekends, split into two groups of 4.

It takes several months to set up citizens’ assemblies (selecting speakers, preparing briefing material, choosing members).  That time will have to be compressed. The Constitution Unit’s small Citizen’s Assembly sent out questionnaires in July 2017 with a view to holding sessions in September and October. That means that an early task for Government and Parliament is to choose who will set up the citizens’ assemblies so that they can get going.

An alternative model for the use of citizens’ assemblies would be to have them on a much smaller scale and right at the start of [each round of] the referendum process. Their rôle would be to set out the key points that need to be addressed in the subsequent debates. This article by Involve, a charity which aims to improve public participation in democracy and which worked with the UCL Constitution Unit reflects on the different options for a citizens’ assembly.



Here is a first cut at an illustration of how that might come together. Much of it is optimistic. It would need a lot of consultation and refinement to come to a workable realistic plan:



Event in Parliament/ Referendum vote

What is happening alongside


Parliamentary resolution sets out key elements of process, including the  options for Round 1



Legislation to launch process completed


Separate legislation on details of process started.



Electoral Commission consults on wording on range of possible options


European Parliament elections


A public body – who? - sets up Citizens’ assemblies




Parliament settles the final wording of the questions to be asked


Process legislation completed

Electoral Commission recommends wording for both rounds of the referendum


A public body – who? - sets up Citizens’ assemblies


Designation of lead campaigners for referendum.




Citizens’ assemblies meet and publish conclusions


Round 1 referendum campaign

25 July

Round 1 of Referendum identifies best available Brexit

The lead campaigner for the winning Brexit option becomes the lead campaigner for Brexit in Round 2



Round 2 referendum campaign



Citizens’ assemblies meet and publish conclusions


Round 2 referendum campaign

Early October

Final Referendum: Brexit or Remain.


If Remain wins: by 31 October

Legislation. Revoke Article 50


If Leave wins: extra time


If decision is for Brexit then we will need to go into extra time for final negotiations and an implementation period.




What use is a time-table anyway?

No-one ever keeps to a time-table. The purpose is to help you to understand what corrective action is needed to get back on course. At this stage, the time-table shows the challenge that Parliament faces to fit everything in.

It should create an incentive for MPs to get on with the decision to have the referendum.


The time-table also confronts MPs with real choices

Looking at the time-table above it may just be too compressed to fit everything in without risking the integrity of the process.

There is not the time to use citizens’ assemblies to help inform the choice of question – though they could still inform voters’ thinking.

There may not be the time to have two voting rounds. That would be problematic for ensuring that all Brexit options were considered by the public and so give a referendum the best chance to settle the question. Having multiple options in a single vote is problematic when some of the options are in radically different categories (Brexit 1, Brexit 2, Remain).

Therefore MPs may feel their only option to stay within the time-table is to have a simple two-choice referendum without add-ons like a prior round, multiple choice questions or citizens’ assemblies. Even that may be too squeezed into the time allowed and lead the Government and MPs to call early for a further short extension.

However, a pared-down referendum would be missing the chance to make the next Brexit referendum a model for conducting referenda. It would also reduce the chance that it would settle the question. So Government and Parliament should consider going back to the EU soon with a plan for a much longer extension to the Spring of 2020 to enable the decision to be made well.



The blogs page of London4Europe is edited by Nick Hopkinson, Vice-Chair. Articles on this page reflect the views of the author and not necessarily of London4Europe.