Background on a Constitutional Convention in the UK

Posted on the 9th March 2016

Contact: Dr Edward Molloy (

For more information, visit


This note first sets out what a citizen-led Constitutional Convention is, and why we need one for the UK. We then flesh out the details of a successful Convention, focusing both on the Convention itself and the public engagement process around it. We concentrate on four potentially controversial areas: remit, outcome, composition and process. Our recommendations are based on in-depth knowledge of international examples and expertise on public engagement in the UK, combined with a recognition of the particular challenges presented by a UK-wide Convention.

Context: what’s a Convention?

A Constitutional Convention is a process for involving members of the public in making decisions about the constitutional shape of a country, region, nation or state.

Conventions and assemblies on constitutional issues have been held in a number of countries and regions, including Ireland, Iceland and British Columbia. The UK also has experience of constitutional conventions, most notably the Scottish Constitutional Convention which paved the way for the creation of the Scottish Parliament.

The design and composition of these conventions reflect the unique geographical, historical and political make up of each of these areas, and the moment in time in which the conventions took place. While there are lessons to be drawn, the UK will need a convention suited to its own distinct composition and needs.

Context: why a Convention now?

There is a public expectation and desire to be more involved in important political decisions, combined with a worrying lack of trust in politics to deliver. To have legitimacy and longevity, the future of devolution in the UK should take its next step with citizens in the forefront. These constitutional issues need to be made with the input and support of citizens across the UK. After all, these issues have to do with things people really care about – their identity, the state of the 2

economy, how public services are run and how the places where people live are shaped and governed. How can citizens not be involved in deciding where power should lie?

Alongside many civil society organisations, faith groups and democracy experts – as well as four of the five largest UK-wide political parties – the Electoral Reform Society is calling for a UK-wide Constitutional Convention: a convention that brings together citizens and politicians with the power to decide how we can renew our politics and improve the way the UK is governed.

In recent years there has been a greater interest, in the UK and elsewhere, in deliberative democracy, where citizens and elected representatives come together to discuss and resolve issues. A UK-wide Constitutional Convention would be an ambitious exercise, probably the largest of its kind so far, with far-reaching potential to demonstrate a different way of doing politics. It is with this in mind that the ERS along with academic partners created a unique experiment in deliberative democracy – the Democracy Matters Citizens’ Assemblies. This allowed us to test different ways of putting citizens at the heart of constitutional decision-making to generate knowledge and capacity to deliver a UK-wide Constitutional Convention.


Setting a specific remit at the outset makes the Convention achievable and understandable, and gives it focus. Something learned from the Democracy Matter project was that the remit must be broad enough to give participants scope to deliberate and have a meaningful input, but not so broad as to make it unfocused.

The distribution of powers between the nations and the regions of the United Kingdom is the most important constitutional question today, the changing nature of which is having a profound effect on how Britain is governed.

In spite of huge public engagement on the issue of devolution in Scotland, there is still a significant gap in the level of debate as to where power should lie in England. The devolved legislatures in Scotland, Wales and Northern Ireland have already had a profound effect on the political culture in these area. The absence of an effectively separate legislature in England has meant that broader discussions of where power should lie have not been as developed there.

It could herefore be preferable to proceed first with an English convention, which addresses the question: “How should England be governed?” This would cover how England relates to the other nations of the UK, and how power can be devolved within England.

An England-only convention would be part of a wider process taking place in the other nations of the UK. There is a clear need for Scottish, Northern Irish and Welsh citizens to have their own constitutional conversations, ideally in a form that allows joint deliberations where appropriate. These can then inform the UK-wide Convention. This will ensure that the unique circumstances are taken into account whilst ensuring that the conversation is a UK-wide one.

Given the current political situation in the UK, we believe there are two questions that the UK-wide Constitutional Convention should address:

  1. How should the nations and regions that are currently within the Union relate to each other?
  2. What are the principles that should underlie the distribution of power in the UK?

Within the scope of (1) are questions around Scottish and Welsh devolution as well as the potential role of an elected second chamber and the role of MPs from devolved regions when it 3

comes to Westminster legislation (including responses to the West Lothian Question). Through (2), decisions can be made about subnational governmental bodies and how power should be devolved to them.

It is good practice for conventions to give participants some scope to go beyond the remit agreed. In Ireland, participants were given extra sessions to cover topics that were not in the original plans but that they felt were important. A similar technique was applied during the Democracy Matters project where it was found to aid concentration on the substantive issues and to heighten a sense of participants’ ‘ownership’ over the process.


It is vital to clarify, from the outset of the process, how the recommendations of the Convention will be taken forward and enacted.

A clear mechanism for enacting the outcomes is the best way to get buy-in from participants and ensure the process is taken seriously. A key lesson from the Scottish independence referendum was that people believe in a process only when they can see their involvement can make a real difference.

The precise nature of the mechanism for taking forward the Convention’s recommendations depends to some degree on the nature of those recommendations and the position the government takes toward any such body. It should be as clear as possible what process they will go through, how the mechanism for taking forward the recommendations will be decided, and how it can be ensured that action will be taken. Some recommendations should return to Parliament for debate followed by a binding referendum.

Given parliamentary sovereignty, Parliament can’t be legally bound to implement the recommendations. But Parliament can commit to honouring the process and enabling the recommendations to be taken forward. (In Ireland, the convention recommended votes at 16 instead of 17 as was suggested on the original agenda). The Government still honoured the process and has committed to a referendum being held).

However, experience of the Democracy Matters project has shown that people can and will engage in a process if they feel that their aspirations and opinions are taken seriously by politicians and political parties. In the absence of a government mandated Constitutional Convention, it would be conceivable that a commitment by all parties who support the principle of a Constitutional Convention to subject the recommendations of such a body to consideration and possible adoption to be implemented in a future parliament. In essence, political buy-in could be created by ascribing to any non-government mandated Convention the power to set the agenda for parties across the political spectrum.


The Convention should be representative of the country, and there are a number of ways of achieving this. We favour a version of the British Columbia citizens’ assembly, owing to its success in gaining fair representation over a large geographic area. The assembly was made up of 160 ‘semi-randomly’ selected citizens. First 100 women and 100 men were selected randomly (stratified within five age groups) from each of the 79 electoral districts (from the electoral register). From those responding to the initial letter, ten men and ten women were invited to attend a local selection meeting in which it was explained what the process involved. From those 4

willing to participate, a random selection of a man and woman for each electoral district was made. This semi-random process (with elements of random and self-selection) enabled the authorities to create a representative sample whilst also ensuring those selected were willing to participate in the substantial workload.

Given England’s size we recommend exploring the possibility of mini-regional assemblies whose recommendations could form part of the England-wide convention.

It is vital that elected representatives have a role and input into the Convention, in order to ensure that the process retains political legitimacy. In Ireland this was achieved by a two-thirds/one-third split on the convention (the majority of participants being citizens). Citizen participants should feel that they are on an equal footing with the politicians, and steps should be taken to instil this culture from the outset.

This approach was piloted in one of the Democracy Matters assemblies and demonstrated that with high quality process design, it was possible for citizens to assess jointly with politicians the relative advantages of various policy positions. It should be recognised however that there are various ways of ensuring that politicians are tied into the process.

Iceland clearly demonstrates the problems of excluding politicians. The process ‘emerged’, rather than being formally designed, and there was a lack of attention to the political reality of what would unfold. A new constitution was not achieved and the failure is largely attributed to the lack of attention to likely political outcomes. Excluding politicians gave them carte blanche to block the process at a later date, and strong citizens’ support came to nothing.

The convention model envisaged here differs from the Scottish Constitutional Convention of the 1990s, in which various civil society groups, politicians, church leaders and the Scottish TUC got together. In a citizen-led model of deliberative democracy, sectional/group interests are not directly represented in the Convention. The Convention process is a deliberative one where decisions are reached through discussion; it is not a negotiation between interested parties.


A proper Convention requires time and commitment. A substantial learning phase (in British Columbia three months (six weekends) were used for the learning phase followed by two months of public hearings) creates the opportunity for participants to gain knowledge and for citizens outside of the Convention to feed into it. Depending on the remit, a comprehensive Convention would need 14-18 months – four-six months for set-up and recruitment.

(The citizens assembly in British Columbia involved a year of work for its members (all meetings took place at weekends). The secretariat was established and selection took place in 2003. Following the selection phase a learning phase ran from January to April 2004 followed by public hearings in May and June. Submissions were invited until September 2004 after which there was a deliberation phase in which the Assembly discussed and debated what its conclusions should be. The final report was presented to the provincial Attorney-General in December 2004)

In Ireland, one weekend was allocated for each of the issues discussed. This was quite restrictive and meant participants had less time to consider the issues, with recommendations being more general as a result. One potential outcome of restricting participants’ learning/deliberating phase is that they may have to rely more heavily on the expert witnesses.

Any Constitutional Convention should engage as wide a cross-section of the public as possible. There are many ways that have been trialled internationally by which the public at large could be 5

enabled to feed into the Convention. These include hosting replica but informal conventions in local areas and the use of online fora to generate and discuss ideas and submissions to the Convention. Membership and grassroots organisations including local political party groups; campaigners, trade unions and faith groups could play a highly significant role by organising their own ‘town hall’ meetings to deliberate before, during and after and formal convention gatherings. Citizens from these meetings could help to set the agenda for the Convention as well as play a direct role by acting as advocates at formal Convention meetings.

A Convention could have two broad phases:

(i) Recruitment of Convention participants and a public engagement process (see below) that allows people outside of the Convention to feed in their views and potentially to help shape the discretionary part of its agenda.

(ii) The second phase will be broken down into a learning section (informed by stage (i)), followed by consultation with constitutional experts, civil society groups and other stakeholders as well as members of the public; then deliberation on the issues; before finally reaching recommendations on these themes/questions. Platforms and techniques such as social media and live-streaming are used to engage members of the public and allow them to follow what’s happening.

Public engagement around a Convention

As already noted, a Constitutional Convention has the potential to combat public disaffection with politics in a way which is high-profile and symbolic. It can show that decision-makers are interested in people’s views and in giving people a significant role in key decisions. To take advantage of this opportunity, significant thought needs to go into how to engage members of the public who are not selected as formal Convention participants.

Based on current best practice of public engagement in the UK and elsewhere, we suggest there are eight criteria that a public engagement process around a Convention should meet. It should be:

  1. Meaningful: Give people a real ability to influence the Convention and be clear with people about the extent of this influence and how it will work
  2. Inclusive: Reach out to diverse communities, ensuring that traditionally excluded groups have their voice heard
  3. Connected: Involve existing national and local networks and community leaders in engaging people with the process
  4. Accessible: Make the process as accessible as possible, for example through realistic timelines, providing a range of simple ways to get involved, being clear about what these are and making it easy to take part
  5. Informed: Provide people with information about relevant issues and options in a balanced way that caters for those with little or no prior knowledge of the subject area
  6. Engaging: Put significant effort into how to engage people, for example via choice of language, tone, messengers, communication channels (both online and offline) and ensuring the process is very visible
  7. Open and transparent: Ensure it is clear at all times how the engagement process and Convention will work, what is happening when the process is live and how decisions are being reached (eg via live streaming, and clear, regular communications)
  8. Followed through: Feedback to people about how their engagement has made a difference and how the outcomes of the Convention are being taken forward, including ensuring that the conclusions of the Convention are fed back in an accessible way. Ideally signpost people to ways in which they can continue their involvement with local and/or national politics

The engagement process would need access to appropriate levels of staff time, budget and expertise to achieve the above.

The Electoral Reform Society is campaigning for a UK-wide Constitutional Convention. We are working cross-party and with a broad range of civil society organisations, academics and faith groups to encourage debate and develop consensus around a model.

Contact: Edward Molloy ( ) For more information, visit

Read more posts...

Automatic Voter Registration (AVR) Briefing

Download PDF (535.4KB pdf) Automatic Voter Registration (AVR) would mean that all eligible voters are put directly on the electoral rolls ensuring that the millions of missing voters are registered and that no one needlessly...

Posted 02 May 2024

Position on mandatory voter ID

Mandatory photographic ID for voters was introduced via the Election Act 2022 and required for the first time in the local elections last year. Voters will be asked to show photographic ID at the local...

Posted 26 Mar 2024

Photo Id sign in polling station