Brexit and the future of Ireland

6.0 SUMMARY

High Court Justice Humphreys in his book ‘Countdown to Unity’ explains how the ‘constitutional imperative’, as outlined by Attorney General Brady (2002-2007), of articles 2 and 3 of the constitution can be achieved by the referendum provided for in Annex A Schedule 1 of the constitutional issues of the Good Friday Agreement.

Attorney General Brady (2002-07) goes on to explain the elements of the Good Friday Agreement and in accepting the Realpolitik of a divided island.

“A fundamental principal of the Good Friday Agreement is that it is a settlement based on the exercise of the right to self-determination by the people of the island of Ireland. The requirement that the right was to be exercised, concurrently, on both parts of the island by way of a separate referendum in each jurisdiction was recognition of the realpolitik of a divided island. The reconciliation of the tension between the right to self-determination and the reality of political life on the island of Ireland is to be found in the policy of consent.”1 Attorney General Brady

In this section we look at Justice Humphreys detailed analysis of the issue of consent, of ‘dual consent’ and the important difference between ‘a’ majority and ‘the’ majority as referred to over the decades by various British Governments. The challenges of a referendum being concurrent in the North & South and how that could and should be interpreted is considered in great detail by Justice Humphreys. The options open to the Irish Government in the event that the Secretary of State for Northern Ireland refuses to hold a referendum or hold a ‘testing the water’ referendum are explored by Justice Humphreys. The triggering of a referendum and its likelihood of being subject to a challenge by way of a referendum petition by unionists is also discussed. The issues surrounding voter fraud in a referendum are outlined. In the event of the referendum being passed the necessity of its ratification by the Irish and British Governments are explained.

The Joint Committee on the Implementation of the Good Friday agreement | 1 Brexit and the future of Ireland

Finally, the research of the Library and Research Service of Leinster House and British House of Commons on the referendums in Quebec and Scotland, where support for separation from a larger political state could not be attained and possible lessons for a referendum here are outlined briefly and the papers are available in full in the appendix at the end of this section.

6.1 RECOMMENDATION Lessons from the failed referendums in Quebec and Scotland need to be learned to ensure that the Irish government fulfils its constitutional obligations of achieving its main aim of the peaceful reunification of Ireland.

2 | The Joint Committee on the Implementation of the Good Friday agreement Brexit and the future of Ireland

6.2 REFERENDUM

“The Good Friday Agreement can be contended to be a permanent feature of the Institutional landscape and to represent a clear road map towards the implementation of a united Ireland,”2 explains Justice Humphries

“The Agreement may be viewed by some as a stable endpoint for political life in Northern Ireland, parking all issues of future unification. However, on another view, the agreement itself expressly recognizes the legitimacy of the drive towards reunification., as a valid political objective to be pursued by peaceful means in accordance with the policy of consent. On that basis, the agreement provides a clear road map towards an act of self-determination by the Irish People (which for this purpose includes British Citizens living in Northern Ireland) that would bring about unity, through the mechanism of an Anglo- Irish Treaty following a referendum vote.”3

The Joint Committee on the Implementation of the Good Friday agreement | 3 Brexit and the future of Ireland

6.2.1 REFERENDUM SOUTH

The Agreement builds on the historic formulation used in the 1994 Joint Declaration by recognizing formally that the constitutional status of Northern Ireland is a matter for the self-determination of the people of the island of Ireland as a whole. This is reflected in paragraph 1(ii) of the constitutional issues section of the Good Friday Agreement,which Provides that:4

CONSTITUTIONAL ISSUES 1. The participants endorse the commitment made by the British and Irish Governments that, in a new British-Irish Agreement replacing the Anglo-Irish Agreement, they will: (i) recognise the legitimacy of whatever choice is freely exercised by a majority of the people of Northern Ireland with regard to its status, whether they prefer to continue to support the Union with Great Britain or a sovereign united Ireland; (ii) recognise that it is for the people of the island of Ireland alone, by agreement between the two parts respectively and without external impediment, to exercise their right of self-determination on the basis of consent, freely and concurrently given, North and South, to bring about a united Ireland, if that is their wish, accepting that this right must be achieved and exercised with and subject to the agreement and consent of a majority of the people of Northern Ireland;5

4 | The Joint Committee on the Implementation of the Good Friday agreement Brexit and the future of Ireland

6.2.2 REFERENDUM IN NORTHERN IRELAND

The constitutional issues section of the agreement provides that Northern Ireland may cease to be part of the UK if a majority so decide in a poll.6

ANNEX A DRAFT CLAUSES/SCHEDULES FOR INCORPORATION IN BRITISH LEGISLATION 1. (1) It is hereby declared that Northern Ireland in its entirety remains part of the United Kingdom and shall not cease to be so without the consent of a majority of the people of Northern Ireland voting in a poll held for the purposes of this section in accordance with Schedule 1.7

The agreement makes clear that the legal mechanics for the holding of a referendum are to be the same as those provided in the Northern Ireland Constitution Act 1973 Schedule 1 (as has been seen, the one previous poll, held under special legislation in 1972, resulted in an overwhelming majority in favour of maintaining of the status quo by reason of the fact that the nationalist community boycotted the poll)8

The Joint Committee on the Implementation of the Good Friday agreement | 5 Brexit and the future of Ireland

6.3 ‘THE’ MAJORITY & ‘A’ MAJORITY

Justice Humphries discusses the concept of dual consent and goes to some lengths to explain the difference between the requirement for ‘the’ majority and ‘a’ majority to vote in favour of a proposal in a future referendum.

Sunningdale Agreement Communique 5. The Irish Government fully accepted and solemnly declared that there could be no change in the status of Northern Ireland until a majority of the people of Northern Ireland desired a change in that status. The British Government solemnly declared that it was, and would remain, their policy to support the wishes of the majority of the people of Northern Ireland. The present status of Northern Ireland is that it is part of the United Kingdom. If in the future the majority of the people of Northern Ireland should indicate a wish to become part of a united Ireland, the British Government would support that wish.

6. The conference agreed that a formal agreement incorporating the declarations of the British and Irish Governments would be signed at the formal stage of the Conference and registered at the United Nations.9

As Humphries explains,

The wording of the commitment of the British government to support unity is also somewhat different to that of the Irish government in the sense that while the Irish government concedes that there can be no change in the status of Northern Ireland until ‘a’ majority of the people of Northern Ireland desire such a change, the wording proposed on behalf of the British government is that a united Ireland would require an expression of wish by ‘the’ majority of the people of Northern Ireland. This is more than a semantic difference in the sense that while ‘a’ majority is a simple test referring to 50 per cent plus one of those participating in a referendum, a requirement for the consent

6 | The Joint Committee on the Implementation of the Good Friday agreement Brexit and the future of Ireland

of ‘the’ majority might be construed as a requirement for consent on behalf of the unionist community as such, a consent which by definition could never be forthcoming as it contradicts the raison d’etre of unionism. The logic of requiring consent of ‘a’ majority is irresistible.10

Humphreys goes on to explain logically

… the really fundamental reason, apart from legal considerations, why a minority or a dual consent could never act to prevent the reunification of the island of Ireland if a majority so wished, is that there is no corresponding provision at present permitting the nationalist and republican ‘minority’ to prevent Northern Ireland from remaining part of the United Kingdom. The test for a United Ireland could not in logic be different from the test for a United Kingdom.11

Oliver Wendell Homes noted it is hard to contend that the tests for a United Kingdom or a united Ireland are different in legal or constitutional terms. In tandem with the concept of ‘dual consent’ (which by and large has emanated from the unionist side of the equation), 12

DUAL CONSENT

It is important to emphasize that both the 1973 Act and crucially the 1998 formula relate to a majority of the electorate who actually turn out to vote in any particular poll. Neither formula requires the consent of an absolute majority of the electorate, or requires consent of the electorate, but subject to a proviso that a certain percentage turns out to vote (similar to a so-called ‘ordinary’ (i.e. non-constitutional) referendum under the Constitution of Ireland). Nor does either test include a requirement that consent be forthcoming from both communities. There is no legal or political justification for the concept of ‘dual consent’ for a united Ireland as there is no such requirement for a United Kingdom.13

Humphreys then states

Neither the 1921 Treaty nor the 1998 Good Friday Agreement contains any provision

The Joint Committee on the Implementation of the Good Friday agreement | 7 Brexit and the future of Ireland

for a minority to veto the basic question of which state the Northern Ireland entity will belong to. There are no vetoes, no requirements for dual consent, and no possibility for individual opt-out by particular counties or areas. The genius of the Good Friday Agreement, it might be contended, is that it permits the ‘majority’ within Northern Ireland effectively to determine which state the Northern Ireland entity will belong to, but permits the ‘minority’ within that entity a very significant share in the public administration of the six counties and a major stake in the orderly government and administration of the entity overall14

The 1985 Agreement did, however, take the UK Government to a legally binding commitment to give effect to the wish of a majority as expressed in a vote for unification a commitment now reflected in the Good Friday Agreement15

GOOD FRIDAY AGREEMENT CONSTITUTIONAL ISSUES Section 1 (ii) recognise that it is for the people of the island of Ireland alone, by agreement between the two parts respectively and without external impediment, to exercise their right of self-determination on the basis of consent, freely and concurrently given, North and South, to bring about a united Ireland, if that is their wish, accepting that this right must be achieved and exercised with and subject to the agreement and consent of a majority of the people of Northern Ireland; 16

8 | The Joint Committee on the Implementation of the Good Friday agreement Brexit and the future of Ireland

6.4 FREELY AND CONCURRENTLY GIVEN:

Turning now to the second problematic issue of the formula used in paragraph 1(ii) of the constitutional issues section, this relates to the precise meaning of the requirement that the exercise of the right of self-determination North and South be ‘freely and concurrently given’. While the requirement that the consent be freely given is relatively unproblematic, the question of concurrent consent does give rise to a difficulty in the matter of timing and form.17

There are a number of difficulties with this analysis having regard to the requirement that the consent of the people of the island of Ireland be given ’concurrently’ in both parts of the island. The difficulty from the point of view of amendment of the constitution is that such an amendment could only have effect in the event of a positive vote in Northern Ireland, a fact that could not be known with certainty in advance. There would seem to be three possible solutions to the question of timing.18

The Joint Committee on the Implementation of the Good Friday agreement | 9 Brexit and the future of Ireland

6.5 A QUESTION OF TIMING: INTERPRETATION OF CONCURRENTLY

In Countdown to Unity there are three possible solutions outlined in relation to the question of timing The first would be not to hold a referendum in Ireland on the same date as a referendum in Northern Ireland but rather to await the result of the Northern Irish referendum and in the event that the result was positive, to then formulate and submit for approval a bill to amend the constitution to give effect to this decision as necessary.

There would seem to be two difficulties with this solution.

Firstly, it could be argued that if the referenda were not held on the same day, the consent would not be concurrent for the purpose of the agreement. Some support for this reading of the agreement can be gathered from the fact that the referenda to approve the agreement itself were envisaged by the agreement as being required to be held on the same day- indeed the date was specified in the agreement itself.

Secondly there is the potential for some delay between the result of a poll in Northern Ireland and a constitutional referendum in Ireland if this was to be postponed until after the holding of such a referendum in Northern Ireland.

Even if the requirement of concurrent consent did not necessarily mean that such consent had to be given on the same day, there might nonetheless be difficulties if the consent was to be given after a very prolonged lapse of time.

A further possible solution to the question of timing would be to hold two referenda in the republic, the first being a plebiscite on the question of unity to be held simultaneously with the Northern Ireland poll, and the second being a subsequent amendment of the constitution to give effect to any constitutional change required on the completion of whatever processes follow from the referendum results North and South. This approach has some attractions from the point of view of logic. However, during the process of the formulating the 1998 constitutional amendment, there was a very strong desire that there would be only a single question put to the people in referendum.

10 | | SenatorThe Joint Mark Committee Daly Report on the for Implementation the Good Friday of Agreement the Good Friday Implementation agreement Committee Brexit and the future of Ireland

Arrangements could have been made in 1998 for the submission to the people of two questions, one of which would have been an approval of the agreement in an identical form of words to that employed in Northern Ireland, and the second would have been the constitutional amendment bill.

However so great was the desire that only a single question be put that the tidy solution of the same question in precisely the same wording being employed both North and South was rejected. It could be contended that the putting of multiple questions does allow a certain amount of what might be termed ‘each-way betting’ by voters who might be dissatisfied with some aspect or another of the proposal. Certainly the methodology used in 1998, which is likely to be of similar relevance in any future referendum, avoids this problem by requiring voters to vote simple yes or no to the totality of the proposal.

A third possibility would be to make whatever amendments to the Constitution are required to give effect to possible future reunification well in advance, in an atmosphere of relative calm, and thus to clear the way for a single simple question to be put in referendum in Ireland on the same date as any poll in Northern Ireland or at least on the same day as a poll likely to result in a ‘yes’ vote. This approach would seem to have advantages on balance, particularly where there may be other constitutional amendments required in advance to facilitate the process of reconciliation and to make the case for a united Ireland.19

It is certainly true that if the secretary of state for Northern Ireland decided to hold a poll in Northern Ireland and if it was decided by the government to hold a poll simultaneously in this part of the island, legislation to facilitate such a poll could be rushed through the Oireachtas on an urgent basis. However, there would seem to be little objective justification for the approach of leaving the matter to the last minute. The preferable approach would seem to be to introduce a bill well in advance of any proposed poll setting out the procedures that would be applied. 20

Any legislation designed to regulate a poll for the purposes of an agreement in Ireland could be closely modelled on the referendum act 1994 with any necessary modifications.21

Senator MarkThe Daly Joint Report Committee for the on Good the Implementation Friday Agreement of the Implementation Good Friday Committeeagreement | 11 Brexit and the future of Ireland

6.6 TESTING THE WATER REFERENDUM

“It should be noted that just as the secretary of state is not under an obligation to make an order for the taking of a poll save where he is of opinion that majority will vote for a united Ireland and no previous poll has been held within the preceding seven years, the Irish Government is not under any express obligation to conduct a simultaneous poll in Ireland. However, given that the purpose of the prevision of the agreement relating to the matter is to vindicate the inherent right of self-determination of the Irish People, it would seem to be an implicit obligation on the Irish Government to hold a simultaneous poll where it was of opinion that the result of the poll in Northern Ireland would be likely to be supportive of a united Ireland.

It could be contended that there would be no such obligation where the Northern Ireland poll was of the ‘testing the water’ variety and not considered on objective evidence likely to result in a change to the ‘status quo’.”22

12 | The Joint Committee on the Implementation of the Good Friday agreement Brexit and the future of Ireland

6.7 REFUSAL TO HOLD A REFERENDUM

Having outlined the mechanism within the Good Friday Agreement by which the referendum to determine the future constitutional status of Northern Ireland would be triggered Justice Humphreys outlines the issues in the event that the Secretary of State for Northern Ireland refuses to hold such a referendum.

The secretary of state is free to refuse to make an order for the holding of a poll unless the conditions laid down in paragraphs 2 and 3 apply ie the secretary of state is of the opinion that a majority would support a united Ireland; and no previous poll had been held during the preceding seven years.23

Of course at present the result of such a poll would be to retain northern Ireland with in the United Kingdom.24

Clearly the decision to hold or not to hold a poll at a time when it is apparent that the result will be a maintenance of the status quo is a matter for political consideration rather than legal obligation, and arguments can be advanced for or against this contention. 25

However, it is clear that if it can be demonstrated to the secretary of state that it is likely that a majority would vote to change the constitutional position of Northern Ireland, then the holding of the poll becomes a mandatory obligation on the secretary of state for Northern Ireland and he/she is required by the agreement to make an order for the purposes of paragraph 1 of schedule 1 to annex A to the constitutional issues section of the agreement.

A perverse subjective refusal to recognise the manifest existence of such a majority would quite possibly have to yield to judicial review on the ground of unreasonableness.

Such a refusal would also amount to a breach of the duty in international law to operate the agreement in good faith (a duty referred to expressly in the 2004 ‘interpretative declaration’ agreed by the two governments ) and to that extent would be liable to be met with international legal proceedings by Ireland26

The Joint Committee on the Implementation of the Good Friday agreement | 13 Brexit and the future of Ireland

6.8 THE INTERNATIONAL COURT OF JUSTICE JURISDICTION TO INCLUDE NORTHERN IRELAND AND THE GOOD FRIDAY AGREEMENT

The Irish Government has become a member of the International Court of Justice since the 15th of December 2011. However there is an exception of any legal dispute with the United Kingdom of Great Britain and Northern Ireland in regard to Northern Ireland Department of Foreign Affairs and Trade

Northern Ireland is not covered by Ireland’s membership of the International Court of Justice and this needs to be addressed. ‘In the absence of such a declaration’ explains Justice Humphreys regarding the Good Friday Agreement are ‘there may simply not be any international judicial venue to which such a complaint could be brought.’27

“The International Court of Justice (ICJ) in The Hague, also known as the World Court, is the principal judicial organ of the United Nations. Although the Court's judgments are binding only on the parties to any particular case, given the calibre of Court's judges and its status as the principal judicial organ of the UN, decisions of the ICJ are themselves often cited as evidence of international law. The Court has given important decisions and opinions on such topics as the law of the sea, boundary disputes, the use of force and the legality of the threat or use of nuclear weapons. It operates under a Statute annexed to the UN Charter. UN Charter and Statute of the ICJ, as published in the Irish Treaty Series.”28

Ireland must fully sign up to the International Court of Justice to protect the Good Friday Agreement.

Of particular concern for the issue of Unity is in the event of the Secretary of State for Northern Ireland not allowing a referendum on a United Ireland when it is believed that a majority would be in favour there is no current recourse for the Irish Government.

Full membership of the International Court of Justice could be a new avenue open to protect the Good Friday Agreement.

14 | The Joint Committee on the Implementation of the Good Friday agreement Brexit and the future of Ireland

6.9 TRIGGERING OF A REFERENDUM

Humphreys argues that

Whether or not there are negotiations to determine the possible shape of a proposal for a united Ireland which might be presented to the people of Northern Ireland in a referendum the trigger for the holding of a referendum is identified in the agreement as being an order made by the secretary of state for Northern Ireland29

While the making of an order is a discretionary matter for the secretary of state he or she is required to make the order in the circumstances set out in paragraph 2 and 3 of schedule 1, as we have seen in the earlier discussion of the 1998 Act.30

SCHEDULE 1 POLLS FOR THE PURPOSE OF SECTION 1 1. The Secretary of State may by order direct the holding of a poll for the purposes of section 1 on a date specified in the order. 2. Subject to paragraph 3, the Secretary of State shall exercise the power under paragraph 1 if at any time it appears likely to him that a majority of those voting would express a wish that Northern Ireland should cease to be part of the United Kingdom and form part of a united Ireland. 3. The Secretary of State shall not make an order under paragraph 1 earlier than seven years after the holding of a previous poll under this Schedule.31

The Joint Committee on the Implementation of the Good Friday agreement | 15 Brexit and the future of Ireland

6.10 REFERENDUM PETITION:

Those legal mechanics for the holding of such a referendum are now well established. Among the features of a referendum in most jurisdictions is the provision for a referendum petition. It can be readily envisaged that the result of a successful referendum to support the proposal for a united Ireland would be likely to be the subject of a referendum petition by some representatives of the unionist minority. Accordingly, it would be of importance to ensure that the referendum itself was carried out entirely in accordance with the governing Northern Ireland legislation and that the grounds on which a referendum petition could be brought were limited to situations where the result was not a fair reflection of the will of the people of Northern Ireland voting in such a poll.32

6.11 VOTER FRAUD

In this context one issue is that of voter fraud but in that regard very significant and stringent legislation on voter identity has been introduced atWestminster33

The Electoral Fraud (Northern Ireland act 2002) appears to have brought this problem under control. Indeed there have been numerous complaints that the problem is now the other way ie that the legislation has inhibited genuine voters from being able to exercise their franchise. None the less the issue of entitlement to vote, electoral registration and electoral fraud will be a key one in the context of polls for the purpose of testing the current strength of public opinion on the constitutional issue. The effect of the electoral legislation, particularly any effect that might discourage qualified voters from exercising their franchise, is in that context a significant question for consideration by the Irish Government and the nationalist political parties in Northern Ireland. 34

16 | The Joint Committee on the Implementation of the Good Friday agreement Brexit and the future of Ireland

6.12 RATIFICATION OF THE DECISION FOR A UNITED IRELAND

The standard procedure adopted to implement an international agreement involving legislative change in two dualist jurisdictions tends to be along lines involving

Firstly signature of a bilateral agreement,

Secondly the introduction and enactment in both jurisdictions of such legislation as may be necessary to implement it,

Thirdly the execution and delivery of instruments of ratification, and

Fourthly the commencement of the agreement either by way of the delivery of the second instrument of ratification or the expiry of a certain period of time from that date, or the happening of some other specified event.

It is likely that the same sequence will apply in the event of an agreement for the handover of Northern Ireland. Indeed the Good Friday Agreement itself envisages firstly an agreement between the two states and secondly the introduction of legislation to give effect to that agreement, with both governments committed to introduce and support in their respective parliaments legislation to give effect to that wish.

Accordingly, following the new British-Irish handover agreement envisaged by the Good Friday Agreement. and the introduction and enactment of appropriate legislation in both the British and Irish parliaments to give effect to that agreement, including if necessary a bill to amend the Constitution which would need to be submitted to referendum if that had not already been provided for by anticipatory amendments to the Constitution, both governments would then execute instruments of ratification and deliver these instruments in a specified manner. The agreement would then commence in accordance with its terms on a specified date, which would also be the date on which the relevant British and Irish legislation would come into effect. The happening of the legal event to trigger such commencement, normally the mechanism specified in the agreement being operated possibly combined with the making of a commencement order for the legislation, would constitute the act bringing into effect the reintegration of the national territory.

The Joint Committee on the Implementation of the Good Friday agreement | 17 Brexit and the future of Ireland

‘Ratification’ of the agreement to transfer Northern Ireland to Irish Sovereignty will take place, in accordance with international law, in the ordinary way by execution of instruments of ratification on each side. The coming into force of the agreement and the formal transfer of Northern Irish will then require revision of political structures on each side but particularly on the Irish Side. In crude terms the British structure will be ‘slimmed down’ by the removal of the Northern Ireland dimension, while the Irish structures will be expanded so that a new thirty-two-county Dail and Seanad will be put in place, the local assembly and executive recognized, local government structures recognised or created for Northern Ireland and provision made for statutory agencies. At its most basic level two new states and parliaments will be constituted or perhaps more accurately, reconstituted, for Ireland and Britain.35

18 | The Joint Committee on the Implementation of the Good Friday agreement Brexit and the future of Ireland

6.13 QUEBEC REFERENDUM 1980 & 1995 LESSONS FOR FUTURE REFERENDUM IN IRELAND ON UNIFICATION.

There have been 2 referendums in Quebec in relation to the issues of economic sovereignty and separation from Canada .

The first on the 20th of May 1980 and the second on the 30th of October 1995, the table below sets out the results of both referendums.

Table 2: Results of Referendums, 1980 and 1995

Referendum 20 May 1980 30 October 1995

Registered Voters 4,367,584 5,087,009

Participation Rate 85.61 93.48

Yes (% of Valid Votes) 40.44 49.42

No (% of Valid Votes) 59.56 50.58

Spoiled Ballots (% of Valid Votes) 1.74 1.82

Source: Gagnon and LaChapelle ,1996 36

The key findings of the Leinster House Library and Research Service for the defeat of the referendum are set out in full in the research paper in the appendix of this section but the main reasons for defeat of the proposals are outlined here . There are some lessons that could be learned from this analysis that can be applied to a campaign to achieve the constitutional imperative of a united Ireland.

The Joint Committee on the Implementation of the Good Friday agreement | 19 Brexit and the future of Ireland

CONCLUSION The 1980 referendum was defeated primarily for the following reasons: · A lack of understanding among voters of the issues relating to sovereignty as well as promises by the federal Canadian government that it would extend greater powers to Quebec after the referendum if Quebec voted No.

By 1995 the Quebec electorate was much more attuned to the issues around sovereignty due firstly to the growing popularity of pro-sovereignty parties such as the PQ and the Bloc Quebecois and the perceived lack of progression with regard to several Canadian governments’ promises to delegate more powers to Quebec. Therefore the margin of defeat in the 1995 referendum was much narrower than that of 1980.

· Nevertheless the referendum was still defeated for a number of reasons, in particular the No campaigns suggestions that, by voting Yes, the Quebec people might lose access to a range of services including their Canadian passports, social welfare and other public service programmes, the right to vote in Canadian federal elections and the loss of some parts of Quebec to native aboriginal peoples as well as the threat of being seen as foreigners by Canada.

· In addition a large rally in Montreal the weekend before the referendum swayed many undecided voters to the Yes camp, in its focus on the continuation of Quebec as a distinct state but within Canada.

20 | The Joint Committee on the Implementation of the Good Friday agreement Brexit and the future of Ireland

6.14 SCOTLAND REFERENDUM 2014

The referendum on the proposal of Scotland leaving the 307 year union with England and Wales was held on the 18th of September 2014.

The result out of over 3 and a half million votes cast was a margin for the No side against the proposal of ‘Should Scotland be an independent country?’ was 10.6%

2,001,926 electors (55.3%) voted ‘No’

1,617,989 electors (44.7%) Voted ‘Yes’37

One of the key findings of the House of Commons Library research paper entitled ‘Scottish Independence Referendum 2014’ was that surveys indicated a clear majority of ‘No’ among women and a very large ‘No’ majority among older voters. Around a fifth of respondents who reported having voted for the SNP at the 2011 Scottish Parliament election voted ‘No’ whereas between a quarter and a third of Labour voters voted ‘Yes’’.

51% of Scottish born respondents voted ‘No’ where as 74% of those born elsewhere in the UK voted ‘No’ as did 59% of those born outside the UK.

The full research paper is at the end of this section in the appendix. There are some lessons that could be learned from this analysis that can be applied to a campaign to achieve the constitutional imperative of a United Ireland

The Joint Committee on the Implementation of the Good Friday agreement | 21 Brexit and the future of Ireland

(ENDNOTES)

1  Richard Humphreys (2009). Countdown to Unity: Debating Irish Reunification. Dublin: Irish Academic Press. P.Xi 2 Ibid. P.6 Introduction 3 Ibid. P. 135 4 Ibid. P.124 5  The Northern Ireland Peace Agreement. Available at http://www.taoiseach.gov.ie/attached_files/Pdf%20files/ NIPeaceAgreement.pdf 6 Richard Humphreys (2009). Countdown to Unity: Debating Irish Reunification. Dublin: Irish Academic Press. P.122 7 Northern Ireland Peace Agreement. 1997. Available at http://www.taoiseach.gov.ie/attached_files/Pdf%20files/NIPeaceAgreement.pdf 8 Richard Humphreys (2009). Countdown to Unity: Debating Irish Reunification. Dublin: Irish Academic Press. P.123 9 The Sunningdale Agreement Communique, 1973. Available at http://cain.ulst.ac.uk/events/sunningdale/agreement.htm 10 Richard Humphreys (2009). Countdown to Unity: Debating Irish Reunification. Dublin: Irish Academic Press. P.42 11 Ibid. P. 7 of Introduction 12 Ibid 13 Ibid. P.41 14 Ibid P.8 15 Ibid P.66 16 Good Friday Agreement. Available at http://www.taoiseach.gov.ie/attached_files/Pdf%20files/NIPeaceAgreement.pdf 17 Richard Humphreys (2009). Countdown to Unity: Debating Irish Reunification. Dublin: Irish Academic Press. P.125 18 Ibid. P.126 19 Ibid P.126,127 20 Ibid. P.128 21 Ibid. P.129 22 Ibid. P.129 23 Ibid. P.121 24 Ibid 25 Ibid. P.122 26 Ibid 27  Richard Humphreys (2009). Countdown to Unity: Debating Irish Reunification. Dublin: Irish Academic Press. P.122

22 | The Joint Committee on the Implementation of the Good Friday agreement Brexit and the future of Ireland

28 Department of Foreign Affairs and Trade. https://www.dfa.ie/our-role-policies/international-priorities/international-law/courts-tribunals-dispute- mechanisms/international-court-of-justice/ 29 Ibid p.120 30 Ibid P.121 31 Good Friday Agreement, 1997. Available at http://www.taoiseach.gov.ie/attached_files/Pdf%20files/NIPeaceAgreement.pdf 32 Richard Humphreys (2009). Countdown to Unity: Debating Irish Reunification. Dublin: Irish Academic Press. P.123 33 Ibid 34 Ibid 35 Ibis. P.145,146 36 Oireachtas Library and Research Paper, The reasons for the defeats of the 1980 and 1995 Referendums in Quebec on sovereignty, January 2017.P.5 37 House of Commons Library, Scottish Independence Referendum 2014, 2014. Available at http://researchbriefings.parliament.uk/ResearchBriefing/Summary/RP14-50

The Joint Committee on the Implementation of the Good Friday agreement | 23 Brexit and the future of Ireland

24 | The Joint Committee on the Implementation of the Good Friday agreement

Research Paper – The reasons for the defeats of the 1980 and 1995 Referendums in Quebec on sovereignty

Research for Deputy Mark Daly

Date: Wednesday, 25th January 2017 Enquiry Number: 2017/00065 Library & Research Service central enquiry desk: Tel – 618 4701

Legal Disclaimer

No liability is accepted to any person arising out of any reliance on the contents of this paper. Nothing herein constitutes professional advice of any kind. For full details of our attribution policy please go to the Library & Research Service’s intranet pages. Please note as per the L&RS 2012 Statement of Service, the L&RS routinely reuses

Oireachtas Library & Research Service | On-Demand Research Paper 0 5847/ Feb 2009

the research it has undertaken for individual Members in order to answer on-demand queries from other Members, or to provide research briefings for all Members.

© Houses of the Oireachtas 2013

Oireachtas Library & Research Service | On-Demand Research Paper 1 5847/ Feb 2009

Introduction

Quebec is one province of ten within the Canadian federation. The question of economic sovereignty for Quebec and separation from Canada has been debated for some time.

Beaulieu…et al. suggest that

“Most observers of the Canadian political scene trace the political instability of Quebec to the creation of the Parti Quebecois (PQ) in 1968, a political party dedicated to Quebec sovereignty.”1

The 1980 referendum on sovereignty was defeated by a margin of almost 20% while the 1995 referendum was narrowly defeated by a margin of just over 1%.

This note sets out the factors behind the defeats of both referendums on sovereignty in Quebec.

Summary of the 1980 and 1995 Referendums in Quebec

1980 Referendum

During the 1976 election campaign the Parti Quebecois (PQ) had indicated that it would hold a referendum on sovereignty. With its winning of that election, the way was clear for a referendum on sovereignty. On November 1, 1979, the Quebec government made public its constitutional proposal in a white paper entitled Québec-Canada: A New Deal. The Québec Government Proposal for a New Partnership Between Equals: Sovereignty-Association.

The referendum sought to ask the people of Québec for a mandate to negotiate, on an equal footing, a new agreement with the rest of Canada which would guarantee Quebec’s sovereignty. The wording is set out below:

1 Beaulieu..[et al.] (2006), p. 624

Oireachtas Library & Research Service | On-Demand Research Paper 2 5847/ Feb 2009

“The Government of Quebec has made public [the white paper referred to above] its proposal to negotiate a new agreement with the rest of Canada, based on the equality of nations. This agreement would enable Quebec to acquire the exclusive power to make its laws, levy its taxes, and establish relations abroad -- in other words, sovereignty -- and at the same time, to maintain with Canada an economic association including a common currency. No change in political status resulting from these negotiations will be effected without approval by the people through another referendum. On these terms, do you agree to give the Government of Quebec the mandate to negotiate the proposed agreement between Quebec and Canada?”

The referendum took place on the 20th of May 1980.

The concept of sovereignty-association was rejected by almost 60 per cent of voters, although it is estimated that about 50 per cent of francophone voters supported it.

Table 1 below provides summary information on the % of the Quebec population identifying as Francophone (mainly or exclusively French-speaking) compared to Anglophones (mainly or exclusively English--speaking) and provides clear evidence of the dominance of French-speaking communities in Quebec.

Table 1: Francophone and Anglophone populations in Quebec % population2 1981 1996 Francophones 82.4% 80.2% Anglophones 10.9% 8.4%

Source: Canadian census, 1981 and 1996

The 1995 Referendum

In preparation for the referendum, every household in Quebec was sent a draft of the Act Respecting the Future of Quebec, with the announcement of the National Commission on the Future of Quebec to commence in February 1995.

The Act Respecting the Future of Quebec (also known as "Bill 1" or the "Sovereignty Bill") was a bill proposed to the Quebec National Assembly by the Parti Québécois

2 % does not add to 100 as there are other linguistic categories included in the Census count

Oireachtas Library & Research Service | On-Demand Research Paper 3 5847/ Feb 2009

government in 1995. It proposed to give the National Assembly the power to declare Quebec "sovereign", with the "exclusive power to pass all its laws, levy all its taxes and conclude all its treaties”. It received a first reading in the National Assembly but the final version of the bill was never voted on following the defeat of the sovereignty option in the 1995 referendum. Had it become law, it would have served as the legal basis for the Quebec government to declare Quebec a sovereign country.

The 1995 Referendum wording was as follows:

“Do you agree that Quebec should become sovereign, after having made a formal offer to Canada for a new Economic and Political Partnership, within the scope of the bill respecting the future of Quebec and of the agreement signed on June 12 1995?”

By means of this referendum wording, the Government of Quebec proposed to the rest of Canada a partnership treaty based on a "Tripartite Agreement" signed on 12 June 1995 between the Parti Quebecois government and 2 smaller pro-sovereignty political parties, the Bloc Québécois and Action Democratique du Quebec.

A yes vote indicated support for negotiations with Canada to lead to a sovereign Quebec.

A no vote indicated continuation of Quebec as a province within a federal Canada.

The referendum was carefully worded by the PQ government to promote the concept of residents of a sovereign Quebec but with the securities provided by being a Canadian citizen.

The campaign began in September 1995 and the referendum was held on the 30th October 1995.

The 1995 Referendum was defeated by less than a 1% margin. Turnout was very high (93.5%) with a large number of spoiled votes (76,000).

In general terms, non-Francophone communities and middle-class voters voted against the Referendum (No vote) while Francophone communities, young people, women, low

Oireachtas Library & Research Service | On-Demand Research Paper 4 5847/ Feb 2009

and lower middle income earners, and the unemployed voted for the Referendum (Yes Vote). 3

Table 2 sets out the results of both referendums

Table 2: Results of Referendums, 1980 and 1995 Referendum 20 May 1980 30 October 1995 Registered Voters 4,367,584 5,087,009 Participation Rate 85.61 93.48 Yes (% of Valid Votes) 40.44 49.42 No (% of Valid Votes) 59.56 50.58 Spoiled Ballots (% of Valid Votes) 1.74 1.82 Source: Gagnon and LaChapelle ,1996

Reasons why the 1980 referendum was defeated

Johnson and McIlraith suggest that in the 1980 referendum language and cultural issues appeared to be more important to voters while economic issues were more important in the 1995 referendum.

The 1980 referendum partly failed because the then Prime Minister of Canada Pierre Trudeau promised Quebec “a new federalism that seemingly promised to satisfy nationalist aspirations within the framework of a revised Canadian constitution”. 4 This emphasis on renewed federalism coupled with Trudeau’s popularity with the electorate in Quebec swung the momentum to the No campaign.

In addition controversial remarks by the Parti Québécois Minister responsible for the Status of Women, which appeared denigrating to female No voters, saw polls slide from 47 per cent in favour of the “Yes” side to 40 per cent by the time of the referendum.

3 ibid, 1996, p. 179

4 Clarke and Kornberg, 1996, p. 677

Oireachtas Library & Research Service | On-Demand Research Paper 5 5847/ Feb 2009

This shows how campaigns tend to matter a lot during referendums because of voter volatility. This issue is addressed in a Spotlight5 issued by the Library and Research Service in 2009.

Reasons why the 1995 referendum was defeated

Opinion polls in the years between the 2 referendums indicated increasing support for sovereignty and the margin of defeat of the Yes vote in the 1995 referendum was much smaller than in the 1980 referendum. LeDuc suggests that most voters had already made up their minds and were more familiar with the issues around sovereignty than were voters in the 1980 referendum6.

Six months before the 1995 referendum, the Yes side was at 39% in the polls but this rose to 50% in some polls during the campaign, and it would appear that the No campaigners underestimated the levels of support for the proposal, initially choosing to remain relatively silent during the early and middle stages of the campaign and only becoming more vocal and public when polls suggested that they might lose the referendum. In addition the sovereignty campaign (the YES vote) had a coherent alliance based around the PQ, the Bloc Quebecois and the small Action Democratique de Quebec. The Bloc Quebecois’ leader Lucien Bouchard was popular among the Francophone electorate and took a leading role in the Yes campaign, becoming the leader of the Oui campaign in the last three weeks of the campaign.

By the time of the 1995 referendum however unemployment was high in Quebec, and fears were stoked by No campaigners about possible higher interest rates and increased taxation should Quebec vote to secede from Canada. Johnson and McIlraith highlight this:

5 http://vhlms-a01/AWData/Library2/Polling_web1.pdf. See part 2, pages 17-20 in particular.

6 LeDuc (1999), p. 20

Oireachtas Library & Research Service | On-Demand Research Paper 6 5847/ Feb 2009

“pro-Canada forces repeatedly issued doomsday forecasts in which a host of misfortunes – political, economic and cultural would befall Quebeckers should they be so foolish as to vote Yes”7

In addition the Canadian Finance Minister suggested in a speech close to the referendum date that Quebeckers would be treated as foreigners in the event of a Yes vote while Canadian Prime Minister Chrétien said that he would take steps toward recognizing Québec as a "distinct society" and guarantee Québec a de facto veto over constitutional changes should it vote No. These late initiatives were designed to appeal to undecided voters and helped swing the campaign to the No side.

Conclusion

The 1980 referendum was defeated primarily for the following reasons:

 A lack of understanding among voters of the issues relating to sovereignty as well as promises by the federal Canadian government that it would extend greater powers to Quebec after the referendum if Quebec voted No.

By 1995 the Quebec electorate was much more attuned to the issues around sovereignty due firstly to the growing popularity of pro-sovereignty parties such as the PQ and the Bloc Quebecois and the perceived lack of progression with regard to several Canadian governments’ promises to delegate more powers to Quebec. Therefore the margin of defeat in the 1995 referendum was much narrower than that of 1980.

 Nevertheless the referendum was still defeated for a number of reasons, in particular the No campaign’s suggestions that, by voting Yes, the Quebec people might lose access to a range of services including their

7 Clarke (2004), p. 349

Oireachtas Library & Research Service | On-Demand Research Paper 7 5847/ Feb 2009

Canadian passports, social welfare and other public service programmes, the right to vote in Canadian federal elections and the loss of some parts of Quebec to native aboriginal peoples as well as the threat of being seen as foreigners by Canada.

 In addition a large rally in Montreal the weekend before the referendum swayed many undecided voters to the Yes camp, in its focus on the continuation of Quebec as a distinct state but within Canada.

Oireachtas Library & Research Service | On-Demand Research Paper 8 5847/ Feb 2009

References

Beaulieu, Marie-Claude, Cosset, Jean-Claude & Essaddam, Naceur, 2006. Political Uncertainty and Stock Market Returns: Evidence from the 1995 Quebec Referendum, The Canadian Journal of Economics, Vol. 39, Issue 2, p. 621-641, accessed at: htttorortale

Clarke, Harold D. & Kornberg, Allan,1996. Choosing Canada? The 1995 Quebec Sovereignty Referendum , PS: Political Science and Politics , Vol. 29, No. 4, pp. 676- 682, accessed at http://www.jstor.org/stable/420791

Clarke, Harold D. & Kornberg, Allan,2004, Referendum Voting as Political Choice: The

Case of Quebec, British Journal of Political Science, Vol. 34, Issue 2, p. 345-355

Gagnon, Alain-G. & Lachapelle, Guy, 1996. Québec Confronts Canada: Two Competing Societal Projects Searching for Legitimacy , Publius Vol. 26, No. 3, pp. 177- 191

House of Commons Library, 2013. The Quebec referendums. Research paper 13/47 accessed at http://researchbriefings.files.parliament.uk/documents/RP13-47/RP13- 47.pdf

Johnson, David R. & McIlwraith, Darren, 1998 Opinion poll and Canadian bond yields during the 1995 Quebec referendum campaign. Canadian Journal of Economics. Vol.

31, Issue 2, p411. Accessed at htttorortale

LeDuc, L., 2002. Referendums and elections: how do campaigns differ?. Routledge ECPR Studies in European Political Science,, 25, pp.145-162. Accessed at https://ecpr.eu/Filestore/PaperProposal/e3c91233-daf8-40a7-896b-969a94eb0826.pdf

Pendergrass, Jan, 2016. “Quebec Referendum of 1995”, Salem Press Encyclopedia, January, 2016.

Oireachtas Library & Research Service | On-Demand Research Paper 9

Scottish Independence Referendum 2014 Analysis of results RESEARCH PAPER 14/50 30 September 2014

In the Scottish Independence Referendum held on the 18th September 2014, the question ‘Should Scotland be an independent country?’ was answered ‘No’ by a margin of 10.6 percentage points.

 2,001,926 electors (55.3%) voted ‘No’

 1,617,989 electors (44.7%) voted ‘Yes’.

This Research Paper presents information on the votes cast and the electorate in each referendum counting area and on opinion polls in the lead-up to the referendum.

Roderick McInnes Steven Ayres Oliver Hawkins

Recent Research Papers

14/39 Small Business, Enterprise and Employment Bill [Bill 11 of 2014-15] 10.07.14 14/40 Unemployment by Constituency, July 2014 16.07.14 14/41 Economic Indicators, August 2014 05.08.14 14/42 Unemployment by Constituency, August 2014 13.08.14 14/43 English local government finance: issues and options 26.08.14 14/44 Pension Schemes Bill [Bill 12 of 2014-15] 21.08.14 14/45 National Insurance Contributions Bill [Bill 80 of 2014-15] 21.08.14 14/46 Economic Indicators, September 2014 02.09.14 14/47 Social Indicators 2014 03.09.14 14/48 International Development (Official Development Assistance Target) 10.09.14 Bill [Bill 14 of 2014-15] 14/49 Unemployment by Constituency, September 2014 17.09.14

Research Paper 14/50 Contributing Authors: Roderick McInnes, Social and General Statistics section Steven Ayres, Social and General Statistics section Oliver Hawkins, Social and General Statistics section

This information is provided to Members of Parliament in support of their parliamentary duties and is not intended to address the specific circumstances of any particular individual. It should not be relied upon as being up to date; the law or policies may have changed since it was last updated; and it should not be relied upon as legal or professional advice or as a substitute for it. A suitably qualified professional should be consulted if specific advice or information is required.

This information is provided subject to our general terms and conditions which are available online or may be provided on request in hard copy. Authors are available to discuss the content of this briefing with Members and their staff, but not with the general public.

We welcome comments on our papers; these should be e-mailed to [email protected].

ISSN 1368-8456

Contents Summary 1

Recent Research Papers 1 Introduction 2

14/39 Small Business, Enterprise and Employment Bill [Bill 11 of 2014-15] 10.07.14 2 Results 3 14/40 Unemployment by Constituency, July 2014 16.07.14 3 Characteristics of local authority areas 8 14/41 Economic Indicators, August 2014 05.08.14 4 Constituency results in Glasgow and Edinburgh 12 14/42 Unemployment by Constituency, August 2014 13.08.14 14/43 English local government finance: issues and options 26.08.14 5 Turnout 13

14/44 Pension Schemes Bill [Bill 12 of 2014-15] 21.08.14 6 Opinion polls 16 14/45 National Insurance Contributions Bill [Bill 80 of 2014-15] 21.08.14 7 Timeline of key events 22 14/46 Economic Indicators, September 2014 02.09.14 14/47 Social Indicators 2014 03.09.14 8 Previous referendums 23 14/48 International Development (Official Development Assistance Target) 10.09.14 Bill [Bill 14 of 2014-15]

14/49 Unemployment by Constituency, September 2014 17.09.14

Research Paper 14/50 Contributing Authors: Roderick McInnes, Social and General Statistics section Steven Ayres, Social and General Statistics section Oliver Hawkins, Social and General Statistics section

This information is provided to Members of Parliament in support of their parliamentary duties and is not intended to address the specific circumstances of any particular individual. It should not be relied upon as being up to date; the law or policies may have changed since it was last updated; and it should not be relied upon as legal or professional advice or as a substitute for it. A suitably qualified professional should be consulted if specific advice or information is required.

This information is provided subject to our general terms and conditions which are available online or may be provided on request in hard copy. Authors are available to discuss the content of this briefing with Members and their staff, but not with the general public.

We welcome comments on our papers; these should be e-mailed to [email protected].

ISSN 1368-8456

RESEARCH PAPER 14/50

Summary In the Scottish Independence Referendum held on the 18th September 2014, the proposition ‘Should Scotland be an independent country?’ was rejected by a margin of 10.6 percentage points.

 2,001,926 electors (55.3%) voted ‘No’

 1,617,989 electors (44.7%) voted ‘Yes’.

There was a majority for ‘No’ in 28 of Scotland’s 32 local authority areas, including the capital, Edinburgh. There was a majority for ‘Yes’ in Scotland’s largest city, Glasgow.

The turnout at the referendum was 84.6% including rejected papers (84.5% based on valid votes). This is the highest turnout at a nationwide referendum or parliamentary election in Scotland since the franchise was extended to women in 1918.

For the first time, 16- and 17-year-olds were eligible to vote. 109,533 people in this age group had registered to vote by polling day.

1 RESEARCH PAPER 14/50

1 Introduction On Thursday 18th September 2014, a referendum was held in Scotland on the question:

“Should Scotland be an independent country?”

The holding of an independence referendum was announced by the Scottish National Party (SNP) after they won an overall majority of seats in the May 2011 Scottish Parliament elections.

The franchise for the referendum was defined by Scottish Independence Referendum (Franchise) Act 2013. Eligibility to vote was based on the franchise at Scottish Parliament and local government elections, which comprises citizens of the UK, the Commonwealth, the Republic of Ireland and other EU member states who are registered electors in Scotland.

A major innovation in this referendum was the lowering of the voting age from 18 to 16 years of age. By 10 March 2014, 98,068 young voters who would be 16 or 17 on the referendum date had already registered to vote.1 The final number of registered young voters in this age bracket as of 18 September 2014 was 109,533 voters2 – accounting for 2.6% of the total referendum electorate and equivalent to around 89% of all 16- to 17-year-olds resident in Scotland.3

Table 1: Scottish referendum electorate – final totals by local authority area

of whom: young of whom: young Total young voters as Total young voters as

electorate voters % of electorate voters % of (16-17) electorate (16-17) electorate

Aberdeen City 175,751 3,658 2.1% Highland 190,787 5,144 2.7% Aberdeenshire 206,490 5,926 2.9% Inverclyde 62,486 1,761 2.8% Angus 93,656 2,490 2.7% Midlothian 69,620 1,747 2.5% Argyll and Bute 72,014 1,808 2.5% Moray 75,173 2,310 3.1% Clackmannanshire 39,974 1,037 2.6% North Ayrshire 113,941 3,377 3.0% Dumfries & Galloway 122,052 2,936 2.4% North Lanarkshire 268,738 7,049 2.6% Dundee City 118,764 3,649 3.1% Orkney Islands 17,806 460 2.6% East Ayrshire 99,682 2,836 2.8% Perth and Kinross 120,052 3,309 2.8% East Dunbartonshire 86,844 2,437 2.8% Renfrewshire 134,745 3,962 2.9% East Lothian 81,947 2,153 2.6% Scottish Borders 95,542 1,798 1.9% East Renfrewshire 72,993 2,479 3.4% Shetland Islands 18,516 544 2.9% Edinburgh, City of 378,039 7,467 2.0% South Ayrshire 94,895 2,646 2.8% Eilean Siar 22,908 598 2.6% South Lanarkshire 261,193 6,246 2.4% Falkirk 122,460 3,007 2.5% Stirling 69,043 2,003 2.9% Fife 302,165 8,155 2.7% West Dunbartonshire 71,128 1,725 2.4% Glasgow City 486,296 10,864 2.2% West Lothian 138,238 3,952 2.9%

Scotland 4,283,938 109,533 2.6% Source: Elections Scotland - results by council and Scottish Independence Referendum Young Voter Registrations

As at 17 September 2014, 789,512 postal votes had been issued (excluding postal proxies).4 These amounted to 18% of the total electorate.

1 General Register Office for Scotland: Electoral Statistics – Scotland 10th March 2014 2 Elections Scotland: Scottish Independence Referendum Young Voter Registrations 3 Based on ONS mid-2013 population estimates. 4 Elections Scotland: Total polling list broken down by local authority 17 Sep 2014

2 RESEARCH PAPER 14/50 RESEARCH PAPER 14/50

1 Introduction 2 Results On Thursday 18th September 2014, a referendum was held in Scotland on the question: Scotland voted No by a margin of 10.6 percentage points (55.3% No; 44.7% Yes).5

“Should Scotland be an independent country?” National result % of valid The holding of an independence referendum was announced by the Scottish National Party Votes cast votes (SNP) after they won an overall majority of seats in the May 2011 Scottish Parliament Yes 1,617,989 44.7% elections. No 2,001,926 55.3% The franchise for the referendum was defined by Scottish Independence Referendum Winning margin for 'No' 383,937 10.6% (Franchise) Act 2013. Eligibility to vote was based on the franchise at Scottish Parliament Total valid votes 3,619,915 100.0% and local government elections, which comprises citizens of the UK, the Commonwealth, the Republic of Ireland and other EU member states who are registered electors in Scotland. Rejected papers 3,429 Total votes cast 3,623,344 A major innovation in this referendum was the lowering of the voting age from 18 to 16 years of age. By 10 March 2014, 98,068 young voters who would be 16 or 17 on the referendum date had already registered to vote.1 The final number of registered young voters in this age bracket as of 18 September 2014 was 109,533 voters2 – accounting for 2.6% of the total Votes cast Vote share referendum electorate and equivalent to around 89% of all 16- to 17-year-olds resident in 2,500,000 3 Scotland. 2,001,926 2,000,000 Table 1: Scottish referendum electorate – final totals by local authority area 1,500,000 1,617,989 of whom: young of whom: young No, Yes, Total young voters as Total young voters as electorate voters % of electorate voters % of 1,000,000 55.3% 44.7% (16-17) electorate (16-17) electorate 500,000

Aberdeen City 175,751 3,658 2.1% Highland 190,787 5,144 2.7% Aberdeenshire 206,490 5,926 2.9% Inverclyde 62,486 1,761 2.8% 0 Angus 93,656 2,490 2.7% Midlothian 69,620 1,747 2.5% Argyll and Bute 72,014 1,808 2.5% Moray 75,173 2,310 3.1% Yes No Clackmannanshire 39,974 1,037 2.6% North Ayrshire 113,941 3,377 3.0% Dumfries & Galloway 122,052 2,936 2.4% North Lanarkshire 268,738 7,049 2.6% Dundee City 118,764 3,649 3.1% Orkney Islands 17,806 460 2.6% The total electorate was 4,283,938. Turnout including rejected papers (the headline figure) East Ayrshire 99,682 2,836 2.8% Perth and Kinross 120,052 3,309 2.8% was 84.6%. Turnout excluding rejected papers (i.e. based on valid votes only) was 84.5%. East Dunbartonshire 86,844 2,437 2.8% Renfrewshire 134,745 3,962 2.9% East Lothian 81,947 2,153 2.6% Scottish Borders 95,542 1,798 1.9% East Renfrewshire 72,993 2,479 3.4% Shetland Islands 18,516 544 2.9% The designated counting areas for the referendum were Scotland’s 32 local authority areas. Edinburgh, City of 378,039 7,467 2.0% South Ayrshire 94,895 2,646 2.8% Eilean Siar 22,908 598 2.6% South Lanarkshire 261,193 6,246 2.4%  There was a majority for ‘No’ in 28 of the 32 areas, including the capital, Edinburgh Falkirk 122,460 3,007 2.5% Stirling 69,043 2,003 2.9% Fife 302,165 8,155 2.7% West Dunbartonshire 71,128 1,725 2.4% (61.1% No). Glasgow City 486,296 10,864 2.2% West Lothian 138,238 3,952 2.9%  The four highest ‘No’ shares were recorded in Scotland’s two northernmost

Scotland 4,283,938 109,533 2.6% authorities (Orkney Islands 67.2% and Shetland Islands 63.7%) and in the two local Source: Elections Scotland - results by council and Scottish Independence Referendum Young Voter Registrations authorities bordering England (Scottish Borders 66.6% and Dumfries and Galloway 65.7%). As at 17 September 2014, 789,512 postal votes had been issued (excluding postal proxies).4 These amounted to 18% of the total electorate.  There was a ‘Yes’ majority in four areas: Dundee (57.3% Yes), West Dunbartonshire (54.0% Yes), Glasgow (53.5% Yes) and North Lanarkshire (51.1% Yes).

1 General Register Office for Scotland: Electoral Statistics – Scotland 10th March 2014 2 Elections Scotland: Scottish Independence Referendum Young Voter Registrations 3 Based on ONS mid-2013 population estimates. 5 Definitive vote counts and electorates for each Scottish local authority were published by the Scottish Electoral 4 Elections Scotland: Total polling list broken down by local authority 17 Sep 2014 Management Board on the Elections Scotland website.

2 3 RESEARCH PAPER 14/50

Table 2.1: results by counting area (in alphabetical order)

total turnout rejec- votes cast (votes cast local authority Yes No yes % no % electorate ted (incl. as % of rejected) electorate)

Aberdeen City 59,390 84,094 180 143,664 41.4% 58.6% 81.7% 175,751 Aberdeenshire 71,337 108,606 102 180,045 39.6% 60.4% 87.2% 206,490 Angus 35,044 45,192 66 80,302 43.7% 56.3% 85.7% 93,656 Argyll and Bute 26,324 37,143 49 63,516 41.5% 58.5% 88.2% 72,014 Clackmannanshire 16,350 19,036 24 35,410 46.2% 53.8% 88.6% 39,974 Dumfries and Galloway 36,614 70,039 122 106,775 34.3% 65.7% 87.5% 122,052 Dundee City 53,620 39,880 92 93,592 57.3% 42.7% 78.8% 118,764 East Ayrshire 39,762 44,442 58 84,262 47.2% 52.8% 84.5% 99,682 East Dunbartonshire 30,624 48,314 73 79,011 38.8% 61.2% 91.0% 86,844 East Lothian 27,467 44,283 48 71,798 38.3% 61.7% 87.6% 81,947 East Renfrewshire 24,287 41,690 44 66,021 36.8% 63.2% 90.4% 72,993 Edinburgh, City of 123,927 194,638 460 319,025 38.9% 61.1% 84.4% 378,039 Eilean Siar 9,195 10,544 19 19,758 46.6% 53.4% 86.2% 22,908 Falkirk 50,489 58,030 107 108,626 46.5% 53.5% 88.7% 122,460 Fife 114,148 139,788 226 254,162 45.0% 55.0% 84.1% 302,165 Glasgow City 194,779 169,347 538 364,664 53.5% 46.5% 75.0% 486,296 Highland 78,069 87,739 168 165,976 47.1% 52.9% 87.0% 190,787 Inverclyde 27,243 27,329 29 54,601 49.9% 50.1% 87.4% 62,486 Midlothian 26,370 33,972 53 60,395 43.7% 56.3% 86.7% 69,620 Moray 27,232 36,935 38 64,205 42.4% 57.6% 85.4% 75,173 North Ayrshire 47,072 49,016 85 96,173 49.0% 51.0% 84.4% 113,941 North Lanarkshire 115,783 110,922 178 226,883 51.1% 48.9% 84.4% 268,738 Orkney Islands 4,883 10,004 20 14,907 32.8% 67.2% 83.7% 17,806 Perth and Kinross 41,475 62,714 96 104,285 39.8% 60.2% 86.9% 120,052 Renfrewshire 55,466 62,067 79 117,612 47.2% 52.8% 87.3% 134,745 Scottish Borders 27,906 55,553 67 83,526 33.4% 66.6% 87.4% 95,542 Shetland Islands 5,669 9,951 15 15,635 36.3% 63.7% 84.4% 18,516 South Ayrshire 34,402 47,247 67 81,716 42.1% 57.9% 86.1% 94,895 South Lanarkshire 100,990 121,800 137 222,927 45.3% 54.7% 85.3% 261,193 Stirling 25,010 37,153 62 62,225 40.2% 59.8% 90.1% 69,043 West Dunbartonshire 33,720 28,776 36 62,532 54.0% 46.0% 87.9% 71,128 West Lothian 53,342 65,682 91 119,115 44.8% 55.2% 86.2% 138,238

Scotland 1,617,989 2,001,926 3,429 3,623,344 44.7% 55.3% 84.6% 4,283,938

4 RESEARCH PAPER 14/50 RESEARCH PAPER 14/50

Table 2.1: results by counting area (in alphabetical order) Table 2.2: results by counting area (ranked by % voting ‘yes’) total turnout total turnout rank rejec- votes cast (votes cast rejec- votes cast (votes cast (1 = local authority Yes No yes % no % electorate local authority Yes No yes % no % ted (incl. as % of ted (incl. as % of highest rejected) electorate) rejected) electorate) % 'yes')

Aberdeen City 59,390 84,094 180 143,664 41.4% 58.6% 81.7% 175,751 Dundee City 53,620 39,880 92 93,592 57.3% 42.7% 78.8% 1 Aberdeenshire 71,337 108,606 102 180,045 39.6% 60.4% 87.2% 206,490 West Dunbartonshire 33,720 28,776 36 62,532 54.0% 46.0% 87.9% 2 Angus 35,044 45,192 66 80,302 43.7% 56.3% 85.7% 93,656 Glasgow City 194,779 169,347 538 364,664 53.5% 46.5% 75.0% 3 Argyll and Bute 26,324 37,143 49 63,516 41.5% 58.5% 88.2% 72,014 North Lanarkshire 115,783 110,922 178 226,883 51.1% 48.9% 84.4% 4 Clackmannanshire 16,350 19,036 24 35,410 46.2% 53.8% 88.6% 39,974 Inverclyde 27,243 27,329 29 54,601 49.9% 50.1% 87.4% 5 Dumfries and Galloway 36,614 70,039 122 106,775 34.3% 65.7% 87.5% 122,052 North Ayrshire 47,072 49,016 85 96,173 49.0% 51.0% 84.4% 6 Dundee City 53,620 39,880 92 93,592 57.3% 42.7% 78.8% 118,764 East Ayrshire 39,762 44,442 58 84,262 47.2% 52.8% 84.5% 7 East Ayrshire 39,762 44,442 58 84,262 47.2% 52.8% 84.5% 99,682 Renfrewshire 55,466 62,067 79 117,612 47.2% 52.8% 87.3% 8 East Dunbartonshire 30,624 48,314 73 79,011 38.8% 61.2% 91.0% 86,844 Highland 78,069 87,739 168 165,976 47.1% 52.9% 87.0% 9 East Lothian 27,467 44,283 48 71,798 38.3% 61.7% 87.6% 81,947 Eilean Siar 9,195 10,544 19 19,758 46.6% 53.4% 86.2% 10 East Renfrewshire 24,287 41,690 44 66,021 36.8% 63.2% 90.4% 72,993 Falkirk 50,489 58,030 107 108,626 46.5% 53.5% 88.7% 11 Edinburgh, City of 123,927 194,638 460 319,025 38.9% 61.1% 84.4% 378,039 Clackmannanshire 16,350 19,036 24 35,410 46.2% 53.8% 88.6% 12 Eilean Siar 9,195 10,544 19 19,758 46.6% 53.4% 86.2% 22,908 South Lanarkshire 100,990 121,800 137 222,927 45.3% 54.7% 85.3% 13 Falkirk 50,489 58,030 107 108,626 46.5% 53.5% 88.7% 122,460 Fife 114,148 139,788 226 254,162 45.0% 55.0% 84.1% 14 Fife 114,148 139,788 226 254,162 45.0% 55.0% 84.1% 302,165 West Lothian 53,342 65,682 91 119,115 44.8% 55.2% 86.2% 15 Glasgow City 194,779 169,347 538 364,664 53.5% 46.5% 75.0% 486,296 Midlothian 26,370 33,972 53 60,395 43.7% 56.3% 86.7% 16 Highland 78,069 87,739 168 165,976 47.1% 52.9% 87.0% 190,787 Angus 35,044 45,192 66 80,302 43.7% 56.3% 85.7% 17 Inverclyde 27,243 27,329 29 54,601 49.9% 50.1% 87.4% 62,486 Moray 27,232 36,935 38 64,205 42.4% 57.6% 85.4% 18 Midlothian 26,370 33,972 53 60,395 43.7% 56.3% 86.7% 69,620 South Ayrshire 34,402 47,247 67 81,716 42.1% 57.9% 86.1% 19 Moray 27,232 36,935 38 64,205 42.4% 57.6% 85.4% 75,173 Argyll and Bute 26,324 37,143 49 63,516 41.5% 58.5% 88.2% 20 North Ayrshire 47,072 49,016 85 96,173 49.0% 51.0% 84.4% 113,941 Aberdeen City 59,390 84,094 180 143,664 41.4% 58.6% 81.7% 21 North Lanarkshire 115,783 110,922 178 226,883 51.1% 48.9% 84.4% 268,738 Stirling 25,010 37,153 62 62,225 40.2% 59.8% 90.1% 22 Orkney Islands 4,883 10,004 20 14,907 32.8% 67.2% 83.7% 17,806 Perth and Kinross 41,475 62,714 96 104,285 39.8% 60.2% 86.9% 23 Perth and Kinross 41,475 62,714 96 104,285 39.8% 60.2% 86.9% 120,052 Aberdeenshire 71,337 108,606 102 180,045 39.6% 60.4% 87.2% 24 Renfrewshire 55,466 62,067 79 117,612 47.2% 52.8% 87.3% 134,745 Edinburgh, City of 123,927 194,638 460 319,025 38.9% 61.1% 84.4% 25 Scottish Borders 27,906 55,553 67 83,526 33.4% 66.6% 87.4% 95,542 East Dunbartonshire 30,624 48,314 73 79,011 38.8% 61.2% 91.0% 26 Shetland Islands 5,669 9,951 15 15,635 36.3% 63.7% 84.4% 18,516 East Lothian 27,467 44,283 48 71,798 38.3% 61.7% 87.6% 27 South Ayrshire 34,402 47,247 67 81,716 42.1% 57.9% 86.1% 94,895 East Renfrewshire 24,287 41,690 44 66,021 36.8% 63.2% 90.4% 28 South Lanarkshire 100,990 121,800 137 222,927 45.3% 54.7% 85.3% 261,193 Shetland Islands 5,669 9,951 15 15,635 36.3% 63.7% 84.4% 29 Stirling 25,010 37,153 62 62,225 40.2% 59.8% 90.1% 69,043 Dumfries and Galloway 36,614 70,039 122 106,775 34.3% 65.7% 87.5% 30 West Dunbartonshire 33,720 28,776 36 62,532 54.0% 46.0% 87.9% 71,128 Scottish Borders 27,906 55,553 67 83,526 33.4% 66.6% 87.4% 31 West Lothian 53,342 65,682 91 119,115 44.8% 55.2% 86.2% 138,238 Orkney Islands 4,883 10,004 20 14,907 32.8% 67.2% 83.7% 32

Scotland 1,617,989 2,001,926 3,429 3,623,344 44.7% 55.3% 84.6% 4,283,938 Scotland 1,617,989 2,001,926 3,429 3,623,344 44.7% 55.3% 84.6%

4 5 RESEARCH PAPER 14/50

Figure 2.1: Map of results by counting area (local authority)

Contains Ordnance Survey data © Crown copyright and database right 2014

6 RESEARCH PAPER 14/50 RESEARCH PAPER 14/50

Figure 2.1: Map of results by counting area (local authority) Figure 2.2: net vote in favour by counting area (local authority) Net % in favour of independence (Yes % minus No %) +20%

+10%

+0%

-10% Scotland average: -10.6%

-20%

-30%

-40% Fife Moray Angus Falkirk Stirling Highland Inverclyde Midlothian Eilean Siar Dundee City Dundee East Lothian East West Lothian West Glasgow City Renfrewshire East Ayrshire East Aberdeen City North Ayrshire Aberdeenshire South Ayrshire South Orkney Islands Orkney Argyll Bute and Scottish Borders Scottish Shetland Islands Shetland North Lanarkshire Perth and Kinross Perth Edinburgh, City of South Lanarkshire South East Renfrewshire East Clackmannanshire East Dunbartonshire East West Dunbartonshire West

Dumfries and Galloway and Dumfries

Net vote in favour (yes votes minus no votes) +40,000

+20,000

+0

-20,000

-40,000

-60,000

-80,000 Fife Moray Angus

Falkirk Stirling Highland Inverclyde Midlothian Eilean Siar Eilean Dundee City Dundee Contains Ordnance Survey data © Crown copyright and database right 2014 Lothian East West Lothian West Glasgow City Renfrewshire East Ayrshire East Aberdeen City North Ayrshire Aberdeenshire South Ayrshire Orkney Islands Orkney Argyll Bute and Scottish Borders Shetland Islands Shetland North Lanarkshire and Kinross Perth Edinburgh, City of South Lanarkshire South East Renfrewshire East

Clackmannanshire East Dunbartonshire East West Dunbartonshire West

Dumfries and Galloway

6 7 RESEARCH PAPER 14/50

3 Characteristics of local authority areas This section illustrates the relationship between the yes vote recorded in each local authority area and the following selection of key demographic and electoral characteristics:

 The SNP’s share of the vote at the May 2014 European Parliament election;

 The percentage of people who describe their national identity as Scottish only, as opposed to both Scottish and British, British only or any other identity (as recorded by the 2011 Census);6

 The percentage of the population born in Scotland;7

 The percentage of the population aged 16-64 who claim an out-of-work benefit (JSA, Income Support or an incapacity-related benefit);8

 The percentage of the population aged 65 and over.9

Of these, out-of-work benefit claimant rate and SNP vote share both appear to show a particularly strong association with the yes-vote share in each area. Scottish-only identity and Scottish birth are also positively correlated with yes-vote share, but not as strongly. By contrast there is a negative correlation, albeit a weak one, between the proportion of the population aged 65-plus in each area and yes-vote share.10

Figure 3.1: yes vote vs SNP vote share at 2014 European election

60%

Dundee City

55% West Dunbartonshire Glasgow City

North Lanarkshire 50% Inverclyde North Ayrshire

Renfrewshire Highland East Ayrshire Eilean Siar Falkirk Clackmannanshire 45% Fife South Lanarkshire West Lothian Midlothian Angus Yes vote Yes South Ayrshire Moray Argyll and Bute Aberdeen City 40% Stirling Perth and Kinross Edinburgh, City of East Dunbartonshire Aberdeenshire East Lothian Shetland Islands East Renfrewshire 35% Dumfries and Galloway Scottish Borders Orkney Islands

30% 15% 20% 25% 30% 35% 40% 45% SNP vote share at the 2014 European Parliament election

6 2011 Census table KS202SC National identity, accessed via Scotland’s Census 2011 7 2011 Census table KS204SC Country of birth, accessed via Scotland’s Census 2011 8 DWP working-age client group data Feb 2014, accessed via ONS Nomisweb 9 ONS mid-2013 population estimates 10 See also Demographic differences and voting patterns in Scotland’s independence referendum, HoC Library Second Reading blog, 23 Sep 2014

8 RESEARCH PAPER 14/50 RESEARCH PAPER 14/50

3 Characteristics of local authority areas Figure 3.2: yes vote vs out-of-work benefit claimant rate This section illustrates the relationship between the yes vote recorded in each local authority 60% area and the following selection of key demographic and electoral characteristics: Dundee City  The SNP’s share of the vote at the May 2014 European Parliament election; 55% West Dunbartonshire

 The percentage of people who describe their national identity as Scottish only, as Glasgow City opposed to both Scottish and British, British only or any other identity (as recorded by the North Lanarkshire 6 50% Inverclyde 2011 Census); North Ayrshire Renfrewshire Highland East Ayrshire 7  The percentage of the population born in Scotland; Eilean Siar Falkirk Clackmannanshire South Lanarkshire 45% West Lothian Fife Angus Midlothian  The percentage of the population aged 16-64 who claim an out-of-work benefit (JSA, vote Yes 8 Moray South Ayrshire Income Support or an incapacity-related benefit); Aberdeen City Argyll and Bute Aberdeenshire Stirling 40% Perth and Kinross  The percentage of the population aged 65 and over.9 Edinburgh, City of East Dunbartonshire East Lothian

East Renfrewshire Of these, out-of-work benefit claimant rate and SNP vote share both appear to show a Shetland Islands 35% particularly strong association with the yes-vote share in each area. Scottish-only identity and Dumfries and Galloway Scottish Borders Scottish birth are also positively correlated with yes-vote share, but not as strongly. By Orkney Islands contrast there is a negative correlation, albeit a weak one, between the proportion of the population aged 65-plus in each area and yes-vote share.10 30% 5% 7% 9% 11% 13% 15% 17% 19% Figure 3.1: yes vote vs SNP vote share at 2014 European election Out-of-work benefit claimant rate (% of population aged 16-64)

60% Figure 3.3: Yes vote vs Census respondents self-describing as Scottish only

Dundee City 60% 55% West Dunbartonshire Dundee City Glasgow City 55% North Lanarkshire West Dunbartonshire 50% Inverclyde Glasgow City North Ayrshire North Lanarkshire Renfrewshire Highland East Ayrshire Eilean Siar Falkirk 50% Inverclyde Clackmannanshire North Ayrshire 45% Fife South Lanarkshire West Lothian Renfrewshire Midlothian Angus Falkirk East Ayrshire Yes vote Yes Highland Moray Clackmannanshire Eilean Siar South Ayrshire South Lanarkshire Argyll and Bute Aberdeen City 45% Fife West Lothian Stirling Perth and Kinross Midlothian 40% vote Yes Angus Edinburgh, City of Aberdeenshire East Dunbartonshire Moray South Ayrshire East Lothian Aberdeen City Argyll and Bute Shetland Islands Stirling Aberdeenshire East Renfrewshire 40% 35% Edinburgh, City of Perth and Kinross East Dunbartonshire Dumfries and Galloway East Lothian Scottish Borders East Renfrewshire Orkney Islands Shetland Islands 35% 30% Dumfries and Galloway Scottish Borders 15% 20% 25% 30% 35% 40% 45% Orkney Islands SNP vote share at the 2014 European Parliament election 30% 45% 50% 55% 60% 65% 70% 75% 2011 Census: % describing their identity as Scottish only 6 2011 Census table KS202SC National identity, accessed via Scotland’s Census 2011 7 2011 Census table KS204SC Country of birth, accessed via Scotland’s Census 2011 8 DWP working-age client group data Feb 2014, accessed via ONS Nomisweb 9 ONS mid-2013 population estimates 10 See also Demographic differences and voting patterns in Scotland’s independence referendum, HoC Library Second Reading blog, 23 Sep 2014

8 9 RESEARCH PAPER 14/50

7 1 2 9 4 8 6 5 3

Figure 3.4: Yes vote vs population born in Scotland 30 26 23 27 17 15 14 31 24 16 32 11 13 20 12 10 28 21 22 19 18 25 29 rank

60%

Dundee City and over 14.8% 17.2% 21.5% 23.4% 17.6% 23.3% 17.2% 18.6% 20.8% 18.8% 18.9% 14.8% 22.9% 17.3% 18.8% 13.9% 20.0% 19.2% 17.9% 19.6% 20.3% 15.8% 21.1% 21.5% 17.9% 22.6% 17.7% 22.9% 17.9% 18.2% 17.3% 14.9% 17.8% % of pop % of

55% Population aged 65 West Dunbartonshire

Glasgow City

8 5 7 1 3 4 6 9 2

North Lanarkshire 28 32 23 17 16 27 18 28 21 20 13 12 24 15 26 30 25 18 31 11 10 21 14 50% rank

RESEARCH PAPER 14/50

Inverclyde North Ayrshire

Eilean Siar Renfrewshire - Highland East Ayrshire Falkirk 7.6% 5.6% 9.4% 7.8% 9.8% 7.6% 9.5% 9.7% 9.3% 8.1% 6.3% 8.4% 9.8% 5.9% 9.5% Clackmannanshire benefit work - 10.3% 14.9% 11.4% 16.0% 15.4% 12.4% 12.5% 18.8% 17.7% 11.9% 17.1% 15.7% 14.3% 13.3% 13.6% 18.1% 12.2% 12.4% of 45% Fife 64 pop.) West Lothian South Lanarkshire - Midlothian Yes vote Yes Moray Angus Out rate (% of 16 of (% rate

South Ayrshire 2014 Feb rate, claimant Aberdeen City Argyll and Bute

Stirling 40% Aberdeenshire Perth and Kinross

Edinburgh, City of East Dunbartonshire 4 9 8 2 7 1 6 5 3

30 22 15 28 14 27 18 10 17 32 16 19 20 26 11 29 25 24 31 21 12 23 13 East Lothian rank

East Renfrewshire Shetland Islands

35% Dumfries and Galloway Census: Scottish Borders

Orkney Islands 75.0% 80.5% 85.9% 76.1% 86.4% 77.1% 83.7% 91.5% 89.2% 83.7% 89.3% 70.2% 85.1% 90.3% 83.6% 82.0% 77.2% 92.9% 88.5% 75.4% 90.4% 93.5% 78.0% 79.3% 91.1% 75.0% 80.7% 87.4% 91.4% 79.8% 92.3% 86.7% 83.3% 2011 2011 % of pop % of born Scotland in

30%

70% 75% 80% 85% 90% 95%

3 5 8 4 6 7 2 9 1

2011 Census: % of population born in Scotland 31 21 11 30 10 24 13 22 17 25 32 16 19 20 27 18 26 12 28 23 15 29 14

rank

%

only identity - 54.7% 61.3% 66.8% 57.4% 67.0% 59.6% 65.5% 70.6% 60.2% 62.6% 59.0% 48.8% 69.2% 68.0% 63.8% 61.9% 61.5% 69.9% 68.3% 58.4% 68.2% 71.6% 62.4% 59.0% 65.9% 57.7% 59.9% 63.9% 67.2% 57.5% 72.0% 65.2% 62.4% Figure 3.5: yes vote vs population aged 65 and over 11 2011 Census: 2011

for local local authorities for Scottish

60%

Dundee City

3 6 2 7 1 5 4 8 9 16 12 22 30 25 26 28 27 23 18 14 21 19 10 13 32 15 29 31 24 17 20 11

55% rank

West Dunbartonshire

Glasgow City % North Lanarkshire year populationyear estimates -

50% Inverclyde 29.6% 31.3% 37.8% 28.4% 33.4% 20.2% 40.3% 33.4% 25.8% 24.4% 22.7% 23.1% 43.1% 33.8% 28.0% 29.2% 30.4% 28.8% 28.9% 33.9% 32.4% 30.8% 16.9% 33.0% 29.6% 20.7% 17.4% 27.1% 29.5% 28.9% 32.2% 32.5% 29.0%

North Ayrshire share, vote SNP Renfrewshire election Euro 2014 East Ayrshire

Highland Falkirk Clackmannanshire Eilean Siar and electoral indicators electoral and

45% West Lothian South Lanarkshire Fife

Midlothian Angus Yes vote Yes

South Ayrshire 1 7 3 9 5 6 4 8 2

Moray 21 24 17 20 12 30 26 27 28 25 10 11 14 16 18 32 23 31 29 19 13 22 15

Aberdeen City Argyll and Bute rank

Stirling

40% Perth and Kinross Edinburgh, City of Aberdeenshire East Dunbartonshire % East Lothian 41.4% 39.6% 43.7% 41.5% 46.2% 34.3% 57.3% 47.2% 38.8% 38.3% 36.8% 38.9% 46.6% 46.5% 45.0% 53.5% 47.1% 49.9% 43.7% 42.4% 49.0% 51.1% 32.8% 39.8% 47.2% 33.4% 36.3% 42.1% 45.3% 40.2% 54.0% 44.8% 44.7% referendum East Renfrewshire demographic

Shetland Islands in share Yes vote 35% Dumfries and Galloway key

Orkney Islands Scottish Borders

30%

selected

13% 15% 17% 19% 21% 23% 25%

:

Percentage of population aged 65 and over, mid-2013

3.1

authority local City Aberdeen Aberdeenshire Angus Bute and Argyll Clackmannanshire Galloway and Dumfries City Dundee Ayrshire East Dunbartonshire East Lothian East Renfrewshire East of City Edinburgh, Siar Eilean Falkirk Fife City Glasgow Highland Inverclyde Midlothian Moray Ayrshire North Lanarkshire North Islands Orkney Kinross and Perth Renfrewshire Borders Scottish Islands Shetland Ayrshire South Lanarkshire South Stirling Dunbartonshire West Lothian West Scotland Table Sources: GRO Scotland 2011 Census; ONS Nomis; ONS 2013 midScotlandSources: 2013 GRO Nomis; ONS ONS 2011 Census;

10 RESEARCH PAPER 14/50

7 1 2 9 4 8 6 5 3

Figure 3.4: Yes vote vs population born in Scotland 30 26 23 27 17 15 14 31 24 16 32 11 13 20 12 10 28 21 22 19 18 25 29 rank

60%

Dundee City and over 14.8% 17.2% 21.5% 23.4% 17.6% 23.3% 17.2% 18.6% 20.8% 18.8% 18.9% 14.8% 22.9% 17.3% 18.8% 13.9% 20.0% 19.2% 17.9% 19.6% 20.3% 15.8% 21.1% 21.5% 17.9% 22.6% 17.7% 22.9% 17.9% 18.2% 17.3% 14.9% 17.8% % of pop % of

55% Population aged 65 West Dunbartonshire

Glasgow City

8 5 7 1 3 4 6 9 2

North Lanarkshire 28 32 23 17 16 27 18 28 21 20 13 12 24 15 26 30 25 18 31 11 10 21 14 50% rank

RESEARCH PAPER 14/50

Inverclyde North Ayrshire

Eilean Siar Renfrewshire - Highland East Ayrshire Falkirk 7.6% 5.6% 9.4% 7.8% 9.8% 7.6% 9.5% 9.7% 9.3% 8.1% 6.3% 8.4% 9.8% 5.9% 9.5% Clackmannanshire benefit work - 10.3% 14.9% 11.4% 16.0% 15.4% 12.4% 12.5% 18.8% 17.7% 11.9% 17.1% 15.7% 14.3% 13.3% 13.6% 18.1% 12.2% 12.4%

45% Fife 64 pop.) West Lothian South Lanarkshire - of Midlothian Yes vote Yes Moray Angus Out rate (% of 16 of (% rate

South Ayrshire 2014 Feb rate, claimant Aberdeen City Argyll and Bute

Stirling 40% Aberdeenshire

Perth and Kinross

Edinburgh, City of East Dunbartonshire 4 9 8 2 7 1 6 5 3

30 22 15 28 14 27 18 10 17 32 16 19 20 26 11 29 25 24 31 21 12 23 13 East Lothian rank

East Renfrewshire Shetland Islands

35% Dumfries and Galloway Census: Scottish Borders

Orkney Islands 75.0% 80.5% 85.9% 76.1% 86.4% 77.1% 83.7% 91.5% 89.2% 83.7% 89.3% 70.2% 85.1% 90.3% 83.6% 82.0% 77.2% 92.9% 88.5% 75.4% 90.4% 93.5% 78.0% 79.3% 91.1% 75.0% 80.7% 87.4% 91.4% 79.8% 92.3% 86.7% 83.3% 2011 2011 % of pop % of born Scotland in

30%

70% 75% 80% 85% 90% 95%

3 5 8 4 6 7 2 9 1

2011 Census: % of population born in Scotland 31 21 11 30 10 24 13 22 17 25 32 16 19 20 27 18 26 12 28 23 15 29 14

rank

%

- only identity 54.7% 61.3% 66.8% 57.4% 67.0% 59.6% 65.5% 70.6% 60.2% 62.6% 59.0% 48.8% 69.2% 68.0% 63.8% 61.9% 61.5% 69.9% 68.3% 58.4% 68.2% 71.6% 62.4% 59.0% 65.9% 57.7% 59.9% 63.9% 67.2% 57.5% 72.0% 65.2% 62.4% Figure 3.5: yes vote vs population aged 65 and over 11 2011 Census: 2011

for local local authorities for Scottish

60%

Dundee City

3 6 2 7 1 5 4 8 9 16 12 22 30 25 26 28 27 23 18 14 21 19 10 13 32 15 29 31 24 17 20 11

55% rank

West Dunbartonshire

Glasgow City % North Lanarkshire year populationyear estimates -

50% Inverclyde 29.6% 31.3% 37.8% 28.4% 33.4% 20.2% 40.3% 33.4% 25.8% 24.4% 22.7% 23.1% 43.1% 33.8% 28.0% 29.2% 30.4% 28.8% 28.9% 33.9% 32.4% 30.8% 16.9% 33.0% 29.6% 20.7% 17.4% 27.1% 29.5% 28.9% 32.2% 32.5% 29.0%

North Ayrshire share, vote SNP Renfrewshire election Euro 2014 East Ayrshire

Highland Falkirk Clackmannanshire Eilean Siar and electoral indicators electoral and

45% West Lothian South Lanarkshire Fife

Midlothian Angus Yes vote Yes

South Ayrshire 1 7 3 9 5 6 4 8 2

Moray 21 24 17 20 12 30 26 27 28 25 10 11 14 16 18 32 23 31 29 19 13 22 15

Aberdeen City Argyll and Bute rank

Stirling

40% Perth and Kinross Edinburgh, City of Aberdeenshire East Dunbartonshire % East Lothian 41.4% 39.6% 43.7% 41.5% 46.2% 34.3% 57.3% 47.2% 38.8% 38.3% 36.8% 38.9% 46.6% 46.5% 45.0% 53.5% 47.1% 49.9% 43.7% 42.4% 49.0% 51.1% 32.8% 39.8% 47.2% 33.4% 36.3% 42.1% 45.3% 40.2% 54.0% 44.8% 44.7% referendum East Renfrewshire demographic

Shetland Islands in share Yes vote 35% Dumfries and Galloway key

Orkney Islands Scottish Borders

30%

selected

13% 15% 17% 19% 21% 23% 25%

:

Percentage of population aged 65 and over, mid-2013

3.1

authority local City Aberdeen Aberdeenshire Angus Bute and Argyll Clackmannanshire Galloway and Dumfries City Dundee Ayrshire East Dunbartonshire East Lothian East Renfrewshire East of City Edinburgh, Siar Eilean Falkirk Fife City Glasgow Highland Inverclyde Midlothian Moray Ayrshire North Lanarkshire North Islands Orkney Kinross and Perth Renfrewshire Borders Scottish Islands Shetland Ayrshire South Lanarkshire South Stirling Dunbartonshire West Lothian West Scotland Table Sources: GRO Scotland 2011 Census; ONS Nomis; ONS 2013 midScotlandSources: 2013 GRO Nomis; ONS ONS 2011 Census;

10 RESEARCH PAPER 14/50

4 Constituency results in Glasgow and Edinburgh In accordance with guidance from the Chief Counting Officer for the Scottish Independence Referendum, official certifications and declarations of referendum results at local level were only made in respect of each local authority area as a whole, and not for wards, constituencies or any other administrative or electoral areas.11

Glasgow and Edinburgh councils have however published constituency breakdowns of the referendum vote in their areas. Glasgow’s votes have been broken down by Scottish Parliament constituency,12 while Edinburgh’s have been disaggregated by UK Parliament constituency.13

Table 4: Glasgow and Edinburgh constituency results

total turnout rejec- votes cast (votes cast Yes No yes % no % ted (incl. as % of rejected) electorate)

Glasgow - total 194,779 169,347 538 364,664 53.5% 46.5% 75.0% Scottish Parliamentary constituencies:

Glasgow Anniesland 23,718 22,976 51 46,745 50.8% 49.2% 79.2% Glasgow Cathcart 26,499 23,688 77 50,264 52.8% 47.2% 80.0% Glasgow Kelvin 23,976 21,742 100 45,818 52.4% 47.6% 68.8% Glasgow Maryhill and Springburn 24,079 18,094 62 42,235 57.1% 42.9% 72.5% Glasgow Pollok 26,807 22,956 45 49,808 53.9% 46.1% 78.7% Glasgow Provan 25,217 19,046 70 44,333 57.0% 43.0% 74.0% Glasgow Shettleston 23,137 21,911 54 45,102 51.4% 48.6% 75.1% Glasgow Southside 21,346 18,934 79 40,359 53.0% 47.0% 71.6%

Edinburgh - total 123,927 194,638 460 319,025 38.9% 61.1% 84.4% UK Parliamentary constituencies:

Edinburgh East 27,500 30,632 100 58,232 47.3% 52.7% 80.7% Edinburgh North and Leith 28,813 43,253 115 72,181 40.0% 60.0% 83.2% Edinburgh South 20,340 38,298 100 58,738 34.7% 65.3% 85.7% Edinburgh South West 24,659 39,509 81 64,249 38.4% 61.6% 84.1% Edinburgh West 22,615 42,946 64 65,625 34.5% 65.5% 88.5%

11 From Elections Scotland Scottish Independence Referendum Guidance for Counting Officers Part E – Verifying and counting the votes, para 7.39: “You must make only one certification and declaration – for your local authority area as a whole and not subdivided into wards, constituencies or any other smaller administrative areas.” 12 Scottish Independence Referendum 2014 Results for Glasgow, Glasgow City Council website 13 Analysis of voting totals in the Scottish Independence Referendum for the City of Edinburgh Area, City of Edinburgh Council website. Constituency electorates provided to the Library.

12 RESEARCH PAPER 14/50 RESEARCH PAPER 14/50

4 Constituency results in Glasgow and Edinburgh 5 Turnout In accordance with guidance from the Chief Counting Officer for the Scottish Independence The turnout at the referendum was 84.6% including rejected papers (84.5% based on valid Referendum, official certifications and declarations of referendum results at local level were votes). This is the highest turnout at a nationwide referendum or parliamentary election in only made in respect of each local authority area as a whole, and not for wards, Scotland since the franchise was extended to women in 1918, surpassing the 81.2% turnout constituencies or any other administrative or electoral areas.11 in Scotland at the 1951 General Election.14

Glasgow and Edinburgh councils have however published constituency breakdowns of the referendum vote in their areas. Glasgow’s votes have been broken down by Scottish Turnout at parliamentary elections and referendums in Scotland since 1918 (%) Parliament constituency,12 while Edinburgh’s have been disaggregated by UK Parliament 0 10 20 30 40 50 60 70 80 90 constituency.13 1918 General Election 1922 General Election Table 4: Glasgow and Edinburgh constituency results 1923 General Election 1924 General Election total turnout 1929 General Election rejec- votes cast (votes cast Yes No yes % no % 1931 General Election ted (incl. as % of 1935 General Election rejected) electorate) 1945 General Election Glasgow - total 194,779 169,347 538 364,664 53.5% 46.5% 75.0% 1950 General Election Scottish Parliamentary constituencies: 1951 General Election 1955 General Election Glasgow Anniesland 23,718 22,976 51 46,745 50.8% 49.2% 79.2% 1959 General Election Glasgow Cathcart 26,499 23,688 77 50,264 52.8% 47.2% 80.0% 1964 General Election Glasgow Kelvin 23,976 21,742 100 45,818 52.4% 47.6% 68.8% 1966 General Election Glasgow Maryhill and Springburn 24,079 18,094 62 42,235 57.1% 42.9% 72.5% 1970 General Election Glasgow Pollok 26,807 22,956 45 49,808 53.9% 46.1% 78.7% 1974 General Election (Feb) 1974 General Election (Oct) Glasgow Provan 25,217 19,046 70 44,333 57.0% 43.0% 74.0% 1975 Referendum, EC membership Glasgow Shettleston 23,137 21,911 54 45,102 48.6% 75.1% 51.4% 1979 Referendum, devolution Glasgow Southside 21,346 18,934 79 40,359 53.0% 47.0% 71.6% 1979 General Election 1979 European Parliament

Edinburgh - total 123,927 194,638 460 319,025 38.9% 61.1% 84.4% 1983 General Election 1984 European Parliament UK Parliamentary constituencies: 1987 General Election Edinburgh East 27,500 30,632 100 58,232 47.3% 52.7% 80.7% 1989 European Parliament Edinburgh North and Leith 28,813 43,253 115 72,181 40.0% 60.0% 83.2% 1992 General Election Edinburgh South 20,340 38,298 100 58,738 34.7% 65.3% 85.7% 1994 European Parliament Edinburgh South West 24,659 39,509 81 64,249 38.4% 61.6% 84.1% 1997 General Election 1997 Referendum, devolution Edinburgh West 22,615 42,946 64 65,625 34.5% 65.5% 88.5% 1999 Scottish Parliament

1999 European Parliament 2001 General Election 2003 Scottish Parliament 2004 European Parliament 2005 General Election 2007 Scottish Parliament 2009 European Parliament 2010 General Election 2011 Scottish Parliament 2014 European Parliament 2014 Referendum, independence

Note: Valid votes as % of electorate. Scottish Parliament turnouts are the higher of the constituency or regional turnout in each case.

11 From Elections Scotland Scottish Independence Referendum Guidance for Counting Officers Part E – Verifying and counting the votes, para 7.39: “You must make only one certification and declaration – for your local authority area as a whole and not subdivided into wards, constituencies or any other smaller administrative areas.” 12 Scottish Independence Referendum 2014 Results for Glasgow, Glasgow City Council website 13 Analysis of voting totals in the Scottish Independence Referendum for the City of Edinburgh Area, City of 14 See HoC Library Research Paper 12/43 UK Election Statistics: 1918-2012 (Aug 2012) and HoC Library Edinburgh Council website. Constituency electorates provided to the Library. Standard Note Elections: Turnout (SN01467)

12 13 RESEARCH PAPER 14/50

Table 5: counting areas ranked by turnout percentage

turnout total turnout rank rejec- votes cast (votes cast local authority yes no yes % electorate (1 = ted (incl. as % of highest % rejected) electorate) turnout)

East Dunbartonshire 30,624 48,314 73 79,011 38.8% 91.0% 86,836 1 East Renfrewshire 24,287 41,690 44 66,021 36.8% 90.5% 72,981 2 Stirling 25,010 37,153 62 62,225 40.2% 90.1% 69,033 3 Falkirk 50,489 58,030 107 108,626 46.5% 88.7% 122,457 4 Clackmannanshire 16,350 19,036 24 35,410 46.2% 88.6% 39,972 5 Argyll and Bute 26,324 37,143 49 63,516 41.5% 88.2% 72,002 6 West Dunbartonshire 33,720 28,776 36 62,532 54.0% 87.9% 71,109 7 East Lothian 27,467 44,283 48 71,798 38.3% 87.6% 81,945 8 Dumfries and Galloway 36,614 70,039 122 106,775 34.3% 87.5% 122,036 9 Scottish Borders 27,906 55,553 67 83,526 33.4% 87.4% 95,533 10 Inverclyde 27,243 27,329 29 54,601 49.9% 87.4% 62,481 11 Renfrewshire 55,466 62,067 79 117,612 47.2% 87.3% 134,735 12 Aberdeenshire 71,337 108,606 102 180,045 39.6% 87.2% 206,486 13 Highland 78,069 87,739 168 165,976 47.1% 87.0% 190,778 14 Perth and Kinross 41,475 62,714 96 104,285 39.8% 86.9% 120,015 15 Midlothian 26,370 33,972 53 60,395 43.7% 86.8% 69,617 16 Eilean Siar 9,195 10,544 19 19,758 46.6% 86.2% 22,908 17 West Lothian 53,342 65,682 91 119,115 44.8% 86.2% 138,226 18 South Ayrshire 34,402 47,247 67 81,716 42.1% 86.1% 94,881 19 Angus 35,044 45,192 66 80,302 43.7% 85.8% 93,551 20 Moray 27,232 36,935 38 64,205 42.4% 85.4% 75,170 21 South Lanarkshire 100,990 121,800 137 222,927 45.3% 85.4% 261,157 22 East Ayrshire 39,762 44,442 58 84,262 47.2% 84.5% 99,664 23 Shetland Islands 5,669 9,951 15 15,635 36.3% 84.4% 18,516 24 North Lanarkshire 115,783 110,922 178 226,883 51.1% 84.4% 268,704 25 North Ayrshire 47,072 49,016 85 96,173 49.0% 84.4% 113,923 26 Edinburgh, City of 123,927 194,638 460 319,025 38.9% 84.4% 378,012 27 Fife 114,148 139,788 226 254,162 45.0% 84.1% 302,165 28 Orkney Islands 4,883 10,004 20 14,907 32.8% 83.7% 17,806 29 Aberdeen City 59,390 84,094 180 143,664 41.4% 81.7% 175,745 30 Dundee City 53,620 39,880 92 93,592 57.3% 78.8% 118,729 31 Glasgow City 194,779 169,347 538 364,664 53.5% 75.0% 486,219 32

Scotland 1,617,989 2,001,926 3,429 3,623,344 44.7% 84.6% 4,283,392

14 RESEARCH PAPER 14/50 RESEARCH PAPER 14/50

Table 5: counting areas ranked by turnout percentage Figure 5.2: Map of turnout by counting area (local authority) turnout total turnout rank rejec- votes cast (votes cast local authority yes no yes % electorate (1 = ted (incl. as % of highest % rejected) electorate) turnout)

East Dunbartonshire 30,624 48,314 73 79,011 38.8% 91.0% 86,836 1 East Renfrewshire 24,287 41,690 44 66,021 36.8% 90.5% 72,981 2 Stirling 25,010 37,153 62 62,225 40.2% 90.1% 69,033 3 Falkirk 50,489 58,030 107 108,626 46.5% 88.7% 122,457 4 Clackmannanshire 16,350 19,036 24 35,410 46.2% 88.6% 39,972 5 Argyll and Bute 26,324 37,143 49 63,516 41.5% 88.2% 72,002 6 West Dunbartonshire 33,720 28,776 36 62,532 54.0% 87.9% 71,109 7 East Lothian 27,467 44,283 48 71,798 38.3% 87.6% 81,945 8 Dumfries and Galloway 36,614 70,039 122 106,775 34.3% 87.5% 122,036 9 Scottish Borders 27,906 55,553 67 83,526 33.4% 87.4% 95,533 10 Inverclyde 27,243 27,329 29 54,601 49.9% 87.4% 62,481 11 Renfrewshire 55,466 62,067 79 117,612 47.2% 87.3% 134,735 12 Aberdeenshire 71,337 108,606 102 180,045 39.6% 87.2% 206,486 13 Highland 78,069 87,739 168 165,976 47.1% 87.0% 190,778 14 Perth and Kinross 41,475 62,714 96 104,285 39.8% 86.9% 120,015 15 Midlothian 26,370 33,972 53 60,395 43.7% 86.8% 69,617 16 Eilean Siar 9,195 10,544 19 19,758 46.6% 86.2% 22,908 17 West Lothian 53,342 65,682 91 119,115 44.8% 86.2% 138,226 18 South Ayrshire 34,402 47,247 67 81,716 42.1% 86.1% 94,881 19 Angus 35,044 45,192 66 80,302 43.7% 85.8% 93,551 20 Moray 27,232 36,935 38 64,205 42.4% 85.4% 75,170 21 South Lanarkshire 100,990 121,800 137 222,927 45.3% 85.4% 261,157 22 East Ayrshire 39,762 44,442 58 84,262 47.2% 84.5% 99,664 23 Shetland Islands 5,669 9,951 15 15,635 36.3% 84.4% 18,516 24 North Lanarkshire 115,783 110,922 178 226,883 51.1% 84.4% 268,704 25 North Ayrshire 47,072 49,016 85 96,173 49.0% 84.4% 113,923 26 Edinburgh, City of 123,927 194,638 460 319,025 38.9% 84.4% 378,012 27 Fife 114,148 139,788 226 254,162 45.0% 84.1% 302,165 28 Orkney Islands 4,883 10,004 20 14,907 32.8% 83.7% 17,806 29 Aberdeen City 59,390 84,094 180 143,664 41.4% 81.7% 175,745 30 Dundee City 53,620 39,880 92 93,592 57.3% 78.8% 118,729 31 Glasgow City 194,779 169,347 538 364,664 53.5% 75.0% 486,219 32

Scotland 1,617,989 2,001,926 3,429 3,623,344 44.7% 84.6% 4,283,392

Contains Ordnance Survey data © Crown copyright and database right 2014

14 15 RESEARCH PAPER 14/50

6 Opinion polls This section examines the trend in referendum voting intentions as measured by opinion polls from January 2012 onwards.15

Figures 6.1 and 6.2 illustrate the trend from January 2012 onwards (including and excluding don’t-knows respectively) while Figures 6.3 and 6.4 focus on the period from January 2014 to referendum day. The trendlines are based on a rolling average of the six most recent opinion polls (each individual set of poll findings is shown as dots).

Measuring the trend on this basis, ‘No’ maintained a substantial average lead over ‘Yes’ in the opinion polls during 2012 and 2013, with a substantial proportion remaining undecided. ‘Yes’ then appeared to narrow the deficit appreciably between February and April 2014 and then almost closed the gap entirely during the final three weeks of the campaign.

Looking at the polls whose last day of fieldwork fell during the final two months of the campaign, the most frequent finding was Yes support of 47%-48% (as against the actual result of 45% - see chart, right).

While this difference lies within the range of the usual +/-3% margin of error for a single poll, this chart indicates a systematic difference between the estimated level of support for Yes and the percentage of people who actually voted that way: 23 of the 29 polls conducted in the last two months of the campaign estimated support for Yes at 46% or more.16 This chimes with research by Prof Stephen Fisher of the University of Oxford who found that opinion polling in the run-up to constitutional referendum votes has tended to overestimate support for the Yes option – for example,:

Looking at 16 recent and/or pertinent constitutional referendums […] In no less than 12 out of the 16 cases the average vote for Yes (which in each case was also the change option) in the final polls was higher than was found in the ballot boxes. The twelve include the referendums on introducing the Alternative Vote in 2011 (5 point difference), Welsh devolution in 2011 (4 point difference) and 1997 (3 points), the Good Friday agreement in 1998 (3 points), Quebec independence in 1995 (4 points)

15 Based on 108 opinion polls published between 29 January 2012 and 17 September 2014, conducted by the following research companies: Panelbase (24 polls); TNS-BMRB (17); YouGov (17); Survation (15); ICM (13); Ipsos MORI (12); Angus Reid (3); Progressive (3); Ashcroft (2); Opinium (2). Poll trackers also maintained by What Scotland Thinks website, the Financial Times and the BBC.Scotland Decides page. See also Polling Data on the Scottish Independence Referendum - Lords Library Note LLN 2014/027, 8 August 2014, for a discussion of polling trends up to Aug 2014. 16 See also So How Well Did The Polls Do?, Prof John Curtice, What Scotland Thinks website, 19 Sep 2014 and How do Scottish referendum polls compare with the result? House of Commons Library Second Reading blog, 19 Sep 2014

16 RESEARCH PAPER 14/50 RESEARCH PAPER 14/50

6 Opinion polls and Scottish devolution in 1979 (3 points). […] the polls overestimated Yes by more than the traditional +/- 3 margin of error in seven of the 16 referendums.17 This section examines the trend in referendum voting intentions as measured by opinion polls from January 2012 onwards.15 Following on from these surveys of prospective voting intention, two polling organisations – YouGov and Lord Ashcroft Polls – conducted retrospective surveys on polling day to ask Figures 6.1 and 6.2 illustrate the trend from January 2012 onwards (including and excluding respondents how they actually voted.18 Both surveys indicated a clear majority for ‘No’ among don’t-knows respectively) while Figures 6.3 and 6.4 focus on the period from January 2014 to women and a very large ‘No’ majority among older voters. Around a fifth of respondents who referendum day. The trendlines are based on a rolling average of the six most recent opinion reported having voted for the SNP at the 2011 Scottish Parliament election voted ‘No’, polls (each individual set of poll findings is shown as dots). whereas between a quarter and a third of Labour voters voted ‘Yes’.

Measuring the trend on this basis, ‘No’ maintained a substantial average lead over ‘Yes’ in Table 6.1: YouGov and Lord Ashcroft referendum-day surveys of Scottish the opinion polls during 2012 and 2013, with a substantial proportion remaining undecided. Independence Referendum voting behaviour ‘Yes’ then appeared to narrow the deficit appreciably between February and April 2014 and selected findings then almost closed the gap entirely during the final three weeks of the campaign. YouGov Lord Ashcroft unweight- unweight- How respondent Yes No How respondent Yes No Looking at the polls whose last ed sample ed sample voted: (%) (%) voted: (%) (%) day of fieldwork fell during the size size final two months of the Overall 2,654 46 54 Overall 2,047 45 55 campaign, the most frequent Gender: Gender: finding was Yes support of Male 1,368 51 49 Male 997 47 53 47%-48% (as against the Female 1,286 42 58 Female 1,050 44 56 actual result of 45% - see chart, right). Age: Age (a): 16-24 216 49 51 25-34 263 59 41 25-39 561 55 45 35-44 384 53 47 While this difference lies within 40-59 1,061 47 53 45-54 415 52 48 the range of the usual +/-3% 60-64 327 45 55 55-64 399 43 57 margin of error for a single poll, 65+ 489 34 66 65+ 488 27 73 this chart indicates a systematic difference between Holyrood 2011 vote Holyrood 2011 vote the estimated level of support Conservative 351 8 92 Conservative 331 2 98 for Yes and the percentage of people who actually voted that way: 23 of the 29 polls Labour 689 27 73 Labour 625 31 69 conducted in the last two months of the campaign estimated support for Yes at 46% or Lib Dem 184 29 71 Lib Dem 263 23 77 SNP 937 78 22 SNP 449 80 20 more.16 This chimes with research by Prof Stephen Fisher of the University of Oxford who Sources: found that opinion polling in the run-up to constitutional referendum votes has tended to YouGov, fieldwork 18 Sep 2014 (non-postal voters), 15 -17 Sep (postal voters) overestimate support for the Yes option – for example,: Lord Ashcroft Polls, fieldwork 18-19 Sep 2014 Note: Figures exclude those that didn’t vote Looking at 16 recent and/or pertinent constitutional referendums […] In no less than 12 (a) Lord Ashcroft poll also found a 71% 'yes' share among 16 -17 year olds and 48% 'yes' among 18-24 year olds, but out of the 16 cases the average vote for Yes (which in each case was also the change based on a sample of only 14 and 84 respondents respectively in these age groups - too small for reliable estimates. option) in the final polls was higher than was found in the ballot boxes. The twelve include the referendums on introducing the Alternative Vote in 2011 (5 point According to the Lord Ashcroft survey, 62% of voters who made their mind up in the final difference), Welsh devolution in 2011 (4 point difference) and 1997 (3 points), the month of the campaign voted ‘Yes’, while 66% of those who had always known their voting Good Friday agreement in 1998 (3 points), Quebec independence in 1995 (4 points) intention voted ‘No’. The overall pattern was that the later a voter made up their mind, the more likely they were to vote ‘Yes’. For yes-voters the most important issues that determined their vote were disaffection with Westminster politics and the NHS, while for no-voters the main issues were the pound and pensions.19

The YouGov survey found that 51% of Scottish-born respondents voted ‘No’, whereas 74% of those born elsewhere in the UK voted ‘No’ as did 59% of those born outside the UK.20 15 Based on 108 opinion polls published between 29 January 2012 and 17 September 2014, conducted by the following research companies: Panelbase (24 polls); TNS-BMRB (17); YouGov (17); Survation (15); ICM (13); Ipsos MORI (12); Angus Reid (3); Progressive (3); Ashcroft (2); Opinium (2). Poll trackers also maintained by What Scotland Thinks website, the Financial Times and the BBC.Scotland 17 Decides page. How accurate will the Scottish independence referendum polls be? Prof Stephen Fisher, What Scotland Thinks See also Polling Data on the Scottish Independence Referendum - Lords Library Note LLN 2014/027, 8 website, 15 Sep 2014 18 August 2014, for a discussion of polling trends up to Aug 2014. YouGov survey, fieldwork 18 Sep 2014 (non-postal voters), 15-17 Sep (postal voters); 16 See also So How Well Did The Polls Do?, Prof John Curtice, What Scotland Thinks website, 19 Sep 2014 and Lord Ashcroft Polls, fieldwork 18-19 Sep 2014 How do Scottish referendum polls compare with the result? House of Commons Library Second Reading blog, 19 Lord Ashcroft Polls, fieldwork 18-19 Sep 2014 19 Sep 2014 20 YouGov survey, fieldwork 18 Sep 2014 (non-postal voters), 15-17 Sep (postal voters)

16 17 RESEARCH PAPER 14/50

Figure 6.1: Scottish independence referendum opinion poll findings, Jan 2012 to Sep 2014: % yes / no / undecided 70 70

65 65

60 60

55 55 no 50 50

45 45

40 40

35 35 yes 30 30

25 25

20 20 undecided 15 15

10 10

5 5

0 0

yes no don't know yes average last 6 polls no average last 6 polls don't know average last 6 polls

18

RESEARCH PAPER 14/50

Figure 6.2: Scottish independence referendum opinion poll findings, Jan 2012 to Sep 2014: % yes / no (excluding undecided) 75 75

70 70

65 65

60 60 no

55 55

50 50

45 45 yes 40 40

35 35

30 30

25 25

yes (excl. don't know) no (excl. don't know) no excl. don't know ave last 6 polls yes excl. don't know ave last 6 polls

19 RESEARCH PAPER 14/50

Figure 6.1: Scottish independence referendum opinion poll findings, Jan 2012 to Sep 2014: % yes / no / undecided 70 70

65 65

60 60

55 55 no 50 50

45 45

40 40

35 35 yes 30 30

25 25

20 20 undecided 15 15

10 10

5 5

0 0

yes no don't know yes average last 6 polls no average last 6 polls don't know average last 6 polls

18

RESEARCH PAPER 14/50

Figure 6.2: Scottish independence referendum opinion poll findings, Jan 2012 to Sep 2014: % yes / no (excluding undecided) 75 75

70 70

65 65

60 60 no

55 55

50 50

45 45 yes 40 40

35 35

30 30

25 25

yes (excl. don't know) no (excl. don't know) no excl. don't know ave last 6 polls yes excl. don't know ave last 6 polls

19 RESEARCH PAPER 14/50

Figure 6.3: Scottish independence referendum opinion poll findings, Jan 2014 to Sep 2014: % yes / no / undecided

60 60

55 55 no 50 50

45 45

40 40

35 35 yes 30 30

25 25

20 20 undecided 15 15

10 10

5 5

0 0

yes no don't know yes average last 6 polls no average last 6 polls don't know average last 6 polls

20

RESEARCH PAPER 14/50

Figure 6.4: Scottish independence referendum opinion poll findings, Jan 2014 to Sep 2014: % yes / no (excluding undecided)

70 70

65 65

60 60 no

55 55

50 50

45 45 yes 40 40

35 35

30 30

yes (excl. don't know) no (excl. don't know) yes excl. don't know ave last 6 polls no excl. don't know ave last 6 polls

21 RESEARCH PAPER 14/50

Figure 6.3: Scottish independence referendum opinion poll findings, Jan 2014 to Sep 2014: % yes / no / undecided

60 60

55 55 no 50 50

45 45

40 40

35 35 yes 30 30

25 25

20 20 undecided 15 15

10 10

5 5

0 0

yes no don't know yes average last 6 polls no average last 6 polls don't know average last 6 polls

20

RESEARCH PAPER 14/50

Figure 6.4: Scottish independence referendum opinion poll findings, Jan 2014 to Sep 2014: % yes / no (excluding undecided)

70 70

65 65

60 60 no

55 55

50 50

45 45 yes 40 40

35 35

30 30

yes (excl. don't know) no (excl. don't know) yes excl. don't know ave last 6 polls no excl. don't know ave last 6 polls

21 RESEARCH PAPER 14/50

7 Timeline of key events 5 May 2011: the Scottish National Party (SNP) wins a majority of seats (69 out of 129) at the Scottish Parliament election and indicates that a referendum on independence will be held in the second half of the parliament.

January 2012: the Scottish and UK Governments launch consultations on the referendum.21 The Scottish Government announces that the referendum will be held in autumn 2014, and proposes that the question should be worded: “Do you agree that Scotland should be an independent country?” It also holds open the possibility of including an extra question on further substantial devolution (or ‘devo-max’).

15 October 2012: the Prime Minister and First Minister sign the ‘Edinburgh Agreement’, enabling the Scottish Government to hold a single-question referendum on independence and to determine the referendum date, franchise and question.22

30 January 2013: the Scottish Government accepts the Electoral Commission’s recommendation that the question be: ‘Should Scotland be an independent country?’23

7 August 2013: the Bill for the Scottish Independence Referendum (Franchise) Act 2013, passed by the Scottish Parliament on 27 June 2013, receives Royal Assent. This bases the referendum franchise on the Scottish Parliament franchise with the addition of 16- and 17- year-olds.

26 November 2013: the Scottish Government publishes its independence White Paper Scotland’s Future.

17 December 2013: the Bill for the Scottish Independence Referendum Act 2013, passed by the Scottish Parliament on 14 November 2013, receives Royal Assent. This sets the referendum date for 18 September 2014.

13 February 2014: in a speech in Edinburgh, Chancellor George Osborne rules out a currency union between the UK and an independent Scotland.24 The Labour Party and Liberal Democrats also express their opposition to a currency union.25

30 May 2014: official 16-week regulated campaign period begins.26

5 August 2014: first televised debate between Alistair Darling and Alex Salmond (on STV)

25 August 2014: second televised debate between Darling and Salmond (on BBC)

2 September 2014: final deadline for voter registration.27

18 September 2014: referendum day.

21 Scotland's constitutional future, Scotland Office, 10 January 2014; Your Scotland – Your Referendum – A Consultation Document, Scottish Government, 25 January 2014 22 Text of Edinburgh Agreement on Gov.uk and on Scottish Government website 23 Government accepts all Electoral Commission recommendations, Scottish Government news release, 30 Jan 2014 24 ‘A currency union with an independent Scotland is ‘not going to happen’ says Chancellor’, Gov.uk 13 Feb 2014 25 ‘Ed Balls: Currency union with Scots ‘won’t happen’’, Scotsman, 13 Feb 2014; Liberal Democrat press release, 13 Feb 2014 26 See Scottish referendum- the campaign rules - Commons Library Standard Note SN06604, 13 May 2014 27 About my vote website

22 RESEARCH PAPER 14/50 RESEARCH PAPER 14/50

7 Timeline of key events 8 Previous referendums 5 May 2011: the Scottish National Party (SNP) wins a majority of seats (69 out of 129) at the The table below summarises referendums that have taken place in the UK since 1973 at a Scottish Parliament election and indicates that a referendum on independence will be held in national or regional level. the second half of the parliament. vote Relevant 21 Subject of referendum Date Outcome share in Turnout January 2012: the Scottish and UK Governments launch consultations on the referendum. area The Scottish Government announces that the referendum will be held in autumn 2014, and favour proposes that the question should be worded: “Do you agree that Scotland should be an NI votes to remain in UK Northern Ireland Northern (note: poll subject to independent country?” It also holds open the possibility of including an extra question on March 1973 98.9% 59% sovereignty referendum Ireland widespread boycott by further substantial devolution (or ‘devo-max’). Nationalist community) Continued membership : the Prime Minister and First Minister sign the ‘ , UK voted to remain in 15 October 2012 Edinburgh Agreement’ of the European UK June 1975 67.2% 64% European Community enabling the Scottish Government to hold a single-question referendum on independence Community and to determine the referendum date, franchise and question.22 Devolution did not Devolution Scotland March 1979 proceed (threshold 51.6% 64% 30 January 2013: the Scottish Government accepts the Electoral Commission’s requirement not met) recommendation that the question be: ‘Should Scotland be an independent country?’23 Devolution did not Devolution Wales March 1979 20.3% 59% proceed Scottish Parliament 7 August 2013: the Bill for the Scottish Independence Referendum (Franchise) Act 2013, Devolution Scotland September 1997 74.3% 60% passed by the Scottish Parliament on 27 June 2013, receives Royal Assent. This bases the established Tax-varying powers for Scottish Parliament given referendum franchise on the Scottish Parliament franchise with the addition of 16- and 17- Scotland September 1997 63.5% 60% year-olds. Scottish Parliament tax-raising powers Welsh Assembly Devolution Wales September 1997 50.3% 50% 26 November 2013: the Scottish Government publishes its independence White Paper established Establishment of Greater London Authority Scotland’s Future. Greater London London May 1998 72.0% 34% established Authority 17 December 2013: the Bill for the Scottish Independence Referendum Act 2013, passed by Northern Support given for Good Friday Agreement May 1998 71.1% 81% the Scottish Parliament on 14 November 2013, receives Royal Assent. This sets the Ireland provisions of Agreement referendum date for 18 September 2014. Establishment of North East Elected Regional elected Regional November 2004 22.1% 47% England Assembly not established 13 February 2014: in a speech in Edinburgh, Chancellor George Osborne rules out a Assembly currency union between the UK and an independent Scotland.24 The Labour Party and Expansion of Welsh Welsh Assembly given Liberal Democrats also express their opposition to a currency union.25 Assembly law making Wales March 2011 greater law making 63.5% 35% powers powers 26 Alternative Vote system 30 May 2014: official 16-week regulated campaign period begins. Voting system UK May 2011 32.1% 42% rejected Scottish Independence 5 August 2014: first televised debate between Alistair Darling and Alex Salmond (on STV) Scotland September 2014 Independence rejected 44.7% 85% Referendum 25 August 2014: second televised debate between Darling and Salmond (on BBC) Sources: 27 House of Lords Constitution Committee - Twelfth Report of session 2009-2010, Referendums in the United Kingdom (7 Apr

2 September 2014: final deadline for voter registration. 2010);

Referendum in Wales - Commons Library standard note SN05897 (Mar 2011)

18 September 2014: referendum day. Alternative Vote Referendum 2011 - Commons Library Research Paper 11/44 (19 May 2014)

There have also been a number of local-authority referendums in England and Wales on the

21 Scotland's constitutional future, Scotland Office, 10 January 2014; Your Scotland – Your Referendum – A question of directly-elected mayors and several authorities have held non-binding 28 Consultation Document, Scottish Government, 25 January 2014 referendums on increases in council tax. 22 Text of Edinburgh Agreement on Gov.uk and on Scottish Government website 23 Government accepts all Electoral Commission recommendations, Scottish Government news release, 30 Jan 2014 24 ‘A currency union with an independent Scotland is ‘not going to happen’ says Chancellor’, Gov.uk 13 Feb 2014 25 ‘Ed Balls: Currency union with Scots ‘won’t happen’’, Scotsman, 13 Feb 2014; Liberal Democrat press release, 13 Feb 2014 26 See Scottish referendum- the campaign rules - Commons Library Standard Note SN06604, 13 May 2014 28 See Commons Library standard notes SN05000 Directly Elected Mayors and SN05682 Council tax: local 27 About my vote website referendums for more details

22 23