Neutral Citation No: [2020] NICA 23 |
Ref: STE11241 | ||
|
|
| |
Judgment: approved by the Court for handing down |
Delivered: 27/04/2020 |
| |
(subject to editorial corrections)* |
|
| |
IN HER MAJESTY’S COURT OF APPEAL IN NORTHERN IRELAND
________
ON APPEAL FROM THE HIGH COURT OF JUSTICE IN NORTHERN IRELAND
QUEEN’S BENCH DIVISION (JUDICIAL REVIEW)
________
Raymond McCord’s Application: Border Poll
_____________
IN THE MATTER OF AN APPLICATION BY RAYMOND McCORD FOR JUDICIAL REVIEW
________
Before: Stephens LJ, Treacy LJ and Colton J
________
STEPHENS LJ (delivering the judgment of the court)
Introduction
[1] There has been longstanding conflict over the partition of Ireland. A key aspiration of Irish nationalists has been to bring about a united Ireland, with the whole island forming one independent state. This aspiration conflicts with that of unionists in Northern Ireland, who want the region to remain part of the United Kingdom. These conflicting aspirations and the resulting conflicts have led to many tragedies with a deep and profoundly regrettable legacy of suffering. A resolution to this aspect of the conflict was found in the Belfast Agreement dated 10 April 1998 which made provision that a united Ireland “must be achieved and exercised with and subject to the agreement and consent of a majority of the people of Northern Ireland.” This means that the status of Northern Ireland will not change without the consent of a majority of its population. The Belfast Agreement also provided that the Secretary of State for Northern Ireland (“the respondent”) was the person responsible for directing by order when a poll is to be held as to whether Northern Ireland shall cease to be a part of the United Kingdom and form part of a united Ireland (“a border poll”). These provisions in the Belfast Agreement were enacted in section 1 and Schedule 1 of the Northern Ireland Act 1998 (“the NIA”). This is not a final outcome between these conflicting aspirations but rather it sets the principles and the mechanism within which the conflict is to be managed. In these judicial review proceedings Raymond McCord (“the appellant”) claimed that there was insufficient clarity and transparency in relation to the mechanism for directing a border poll. He claimed that the refusal or failure of the respondent to have a policy setting out the circumstances in which he will direct the holding of a border poll is a breach of the constitutional issues provided for in the Belfast Agreement. He sought an order that the respondent publish a policy setting out the circumstances in which he will direct the holding of a border poll. The issues for determination at first instance were (a) whether the respondent was required to publish a policy governing the discretion to hold a border poll under section 1 and Schedule 1 paragraph 1 of the NIA and (b) whether the respondent was required to publish a policy governing the requirement to hold a border poll under section 1 and Schedule 1 paragraph 2 of the NIA.
[2] The appellant’s application for judicial review was dismissed by Sir Paul Girvan (“the judge”) under citation [2018] NIQB 106 and the appellant brings this appeal.
[3] At the date of the first instance judgment the Secretary of State for Northern Ireland was the Rt Hon Karen Bradley, M.P. The present Secretary of State for Northern Ireland is the Rt Hon Brandon Lewis CBE, M.P.
[4] Mr Lavery QC appeared with Mr Fegan for the appellant. Dr McGleenan QC appeared with Mr McLaughlin and Ms Ellison on behalf of the respondent.
The first instance proceedings
[5] We consider it appropriate to set out a sequence in relation to the first instance proceedings to identify precisely the issues which were before the judge and precisely which issues are before this court.
[6] On 21 June 2017 the appellant applied ex parte pursuant to Order 53, Rule 3 of the Rules of the Court of Judicature (Northern Ireland) 1980 for leave to apply for judicial review. On the same date the applicant lodged his Order 53 statement and an affidavit. The relief sought by the appellant all centred around “the respondent’s failure or refusal to have a policy which sets out the circumstances in which the respondent will order the holding of a (border) poll.” At this stage the appellant was not making the case that there had been a decision by the Secretary of State not to hold a border poll (“a negative decision”). No case was being made that if there had been a negative decision then that decision was unlawful. Furthermore, at this stage the appellant was not seeking an order of certiorari quashing the decision not to hold a border poll. Finally, in that part of the Order 53 statement which sought “particulars of any claim to rights under the European Convention on Human rights” (“ECHR”) the appellant answered “none.”
[7] The respondent had been notified of the application for leave to apply for judicial review and had been permitted by the judge to make submissions in relation to that application.
[8] The appellant lodged a skeleton argument dated 3 September 2017 in relation to the application for leave. Again, in that skeleton argument there was no suggestion of any challenge to an alleged negative decision.
[9] In a proposed amended Order 53 statement dated 6 October 2017 the appellant set out amended grounds which included reference to the respondent’s “decision not to hold a (border) poll.” The appellant sought a declaration that the decision not to hold a border poll was unlawful. He also sought an order of certiorari quashing the decision not to hold a border poll. Other proposed amendments to the grounds on which relief was sought included for instance:
“(l) … the respondent has unlawfully fettered his discretion to hold a (border) poll under para 1 Schedule 1 NIA.
(m) The respondent has failed to take into account material considerations namely:
(i) the result of an opinion poll which shows support for holding a section 1 NIA (border) poll;
(ii) The political stability which may flow from a clear vote in favour of maintaining the Union.”
Again, in that part of the proposed amended Order 53 statement which sought “particulars of any claim to rights under the (ECHR)” the appellant answered “none.”
[10] The proposed amended Order 53 statement was followed on 20 October 2017 by a further appellant’s skeleton argument which under the heading “Error of Law” at paragraph [42] included arguments in relation to further grounds of challenge. For instance, an argument that “by asserting that the conditions for a referendum to be held on Northern Ireland’s constitutional status have not been met, the Secretary of State has fettered his discretion on the exercise of his otherwise unfettered discretion” under Schedule 1 paragraph 1 of the NIA. Another instance is that “by stating in the pre-action response and in other public statements that it is his clear view that a majority of the people of Northern Ireland continue to support the current political settlement including Northern Ireland’s position within the United Kingdom, the Secretary of State has made an error of law and misdirected himself as the circumstances within which he is to hold a poll under Schedule 1 paragraph 2 of the NIA. The error of law was said to be that “the statutory test under paragraph 2 is that ‘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’ not that they ‘continue to support the current political settlement.’” It was also argued on behalf of the appellant that “as the current political settlement includes the Belfast Agreement and the standing right of the people of Northern Ireland to self‑determination under it, this position is circular and meaningless. In considering the test in this manner, the respondent has made an error of law/misdirected himself and/or taken into account an irrelevant consideration.”
[11] In response to the application for leave to apply for judicial review and also in response to the proposed amended Order 53 statement the respondent lodged the affidavit sworn on 16 January 2018 of Ruth Sloan, Deputy Director, Political Strategy and Implementation Group of the Northern Ireland Office. At paragraph [11] of that affidavit Ms Sloan asserted that the pre-action response did not “convey that the Secretary of State had taken a formal negative decision not to hold a border poll.”
[12] On 1 March 2018 the judge directed that a leave hearing be held.
[13] In advance of that hearing there was a further appellant’s skeleton argument dated 7 March 2018 and a respondent’s skeleton argument dated 11 March 2018.
[14] At the leave hearing on 13 March 2018 the judge in an ex tempore ruling set out how the grounds upon which the appellant had relied had progressed as between the original Order 53 statement and the proposed amended Order 53 statement. He referred to the respondent’s proposition “that no decision as such has been made in relation to the discretionary power to hold a border poll.” The judge stated that the case “put forward by the respondent is that the Secretary of State has not seen anything that caused for the need to exercise the discretionary power to have a border poll and the argument is that effectively no decision as such has been made to hold a poll and therefore there is no reviewable decision on the part of the Secretary of State.” The judge also noted the appellant’s counter argument that “in effect the Secretary of State in not calling a border poll must have made the decision not to call a poll and that in itself is a reviewable decision ….” The judge concluded that he was not “persuaded that a decision has been made which is reviewable … not to hold a (border) poll.” He granted leave to apply for judicial review on the issues as to whether the respondent’s failure or refusal to have a policy which sets out the circumstances in which the respondent will order the holding of a border poll was unlawful. He refused leave to apply for judicial review in relation to the proposed amended grounds in the Order 53 statement dated 6 October 2017. He identified the result of the leave hearing as being “that the case will proceed with leave being granted in relation to (the) policy issue. I will not allow the amendments to the Order 53 Statement to extend the case beyond that and the case will proceed accordingly to a substantive hearing on that issue.”
[15] The appellant did not seek to appeal the decision of the judge refusing leave to apply for judicial review in relation to any of the proposed amendments and refusing to amend the Order 53 statement.
[16] The first instance hearing proceeded on the basis of the Order 53 Statement dated 21 June 2017.
[17] After the first instance judgment was delivered the appellant lodged a Notice of Appeal dated 3 August 2018. The first three grounds of appeal and also the tenth ground raised matters in relation to which the judge had refused leave to amend or to apply for judicial review. For instance the first ground was that:
“The learned judge erred in fact and law in refusing leave for judicial review on the ground that the respondent had made an error of law and/misdirected herself in law because her decision not to hold a poll failed to take into account the discretion she has to hold a poll under paragraph 1 of Schedule 1 NIA.”
The inclusion of these grounds of appeal are to be seen in the context that the judge had refused the proposed amendment; refused leave in relation to them; they had not been advanced or adjudicated upon at first instance; there had been no appeal within time against the judge’s refusal to amend or to grant leave nor any application to extend time in relation to the appeal contained in grounds 1-3 and 10.
[18] At the hearing of the appeal Mr Lavery stated that the appellant was no longer relying on grounds 1-3 and 10 in the Notice of Appeal. He also stated that this court should ignore paragraph [51] of the appellant’s skeleton argument dated 15 December 2019 which was an exact copy of paragraph [42] of the skeleton argument dated 20 October 2017 in respect of which leave had been refused. Mr Lavery also deleted paragraph [5](b) of the appellant’s skeleton argument dated 15 December 2019 which also referred to “the decision not to hold a border poll.”
[19] Paragraphs [10], [32] and [47] of the appellant’s skeleton argument dated 15 December 2019 included a submission that by virtue of the Legislative and Regulatory Reform Act 2006 the respondent was required to have and publish a policy in that he was carrying out a regulatory function under section 1 and schedule 1 NIA. The appellant relied upon the statutory obligation under section 21 of the 2006 Act that in carrying out “regulatory functions” regard must be had to the principles of accountability, transparency, proportionality and consistency. It was said that a policy could be the only manner of having regard to those principles. This was not a ground in the Order 53 Statement or in the proposed amended Order 53 Statement or in the Notice of Appeal. It had not been advanced or adjudicated upon at first instance. There was no application to amend the Notice of Appeal to include this further ground.
[20] In the respondent’s skeleton argument Dr McGleenan objected to the appellant relying on the Legislative and Regulatory Reform Act 2006 but in the alternative submitted that this new point was entirely without merit. On behalf of the respondent it was submitted that:
(i) Section 21 does not lay down a principle of general application to all statutory powers. Pursuant to section 24 of the 2006 Act, the principles apply only to those functions which have been designated by a Minister of the Crown pursuant to a Statutory Order. Several Orders have been made under the 2006 Act, which define the functions to which section 21 applies. None of them include the powers under section 1 and Schedule 1 NIA. Accordingly, it was submitted that section 21 of the 2006 Act was of no relevance to this appeal.
(ii) The exercise of the power to hold a border poll is not a “regulatory function” within the meaning of section 32(2) of the 2006 Act since it does not involve the imposition of a requirement, restriction, condition or the setting of a standard. It involves making an assessment of political opinion and the public interest in Northern Ireland, for constitutional purposes. This it was submitted was supported by the Explanatory Notes for this provision, which states:
“165. The Act does not attempt to define a “regulator”, but rather provides a broad definition of the functions carried out by such persons or bodies. Functions falling within the definition might be exercised by a wide range of bodies including Government departments, local authorities and independent statutory regulators. The first limb of the definition (subsection (2)(a)) is aimed at functions of ‘regulating’ (for example by producing rules, or imposing requirements, which apply to a category of persons). The second limb (subsection (2)(b)) covers functions of enforcing or securing compliance with such regulation. A regulatory function of making rules or regulations falling within the first limb could be exercised by a different person than the corresponding regulatory function (falling within the second limb of the definition) of securing compliance with or enforcing those rules or regulations.”
[21] In the light of that response Mr Lavery did not seek leave to amend the Notice of Appeal nor did he advance any argument based on the Legislative and Regulatory Reform Act 2006.
[22] Another submission contained at paragraph [34] of the appellant’s skeleton argument dated 15 December 2019 was that the respondent was under a duty to consider whether he should exercise discretion to direct the holding of a border poll. Again this was not a ground in the Order 53 Statement; in the proposed amended Order 53 Statement; it had not been advanced or adjudicated upon at first instance and it was not a ground in the Notice of Appeal. Mr Lavery did not rely on it at the hearing in this court.
[23] In conclusion we proceed on the basis of the issues contained in the Order 53 Statement dated 21 June 2017 and grounds 4-9 and 11 in the appellant’s Notice of Appeal. We note that the Order 53 statement expressly states that no claim to rights under the ECHR are relied upon by the appellant.
The appellant’s submissions at first instance and in this court
[24] We summarise the appellant’s submissions both at first instance and in this court.
[25] Inconsistency. Under this heading it was submitted that the requirement to have a policy is a straightforward one as “inconsistency is a ground for judicial review.” It was said that a policy promotes consistency and that as a general proposition of law there is a need, or indeed, a duty for a decision-maker to have a policy in the exercise of their powers particularly where, as here the discretionary power is a wide one and where the exercise of those powers is of gravity or significance. The appellant relied on a number of authorities to support this ground of challenge including R (on the application of Alconbury Developments Ltd) v Secretary of State for the Environment, Transport and the Regions [2003] 2 AC 295 at paragraph [143]; In Re Findlay [1985] 1 AC 318 at page 335 letters D-H; and R v Secretary of State for the Home Department, Ex Parte Venables [1997] 2 WLR 67 at page 90 letter B.
[26] Policy and the rule of law. It was submitted that there has been recognition of the connection between having a policy and the rule of law. It was said that as a consequence the rule of law requires there to be a policy governing directions as to the holding of a border poll. The authorities relied on by the appellant in support of this proposition were R (L and another) v Secretary of State for the Home Department [2003] 1 WLR 1230 at paragraph [25] letter G; R (on the application of Lumba) v Secretary of State for the Home Department [2011] 4 All ER 1 at paragraphs [30] - [34] and [302]; B v Secretary of State for Work and Pensions [2005] EWCA Civ 929 at paragraph [43]; and Re Rodger's Application [2014] NIQB 79 at paragraph [86].
[27] Transparency and the requirement for an adequate policy. It was submitted that the principle of transparency having evolved out of Strasbourg jurisprudence was now an established constitutional and legal principle. It was submitted that transparency was now a component of the ‘rule of law’ in accordance with the principle of ‘legal certainty.’ It was also submitted that “where there is or ought to be a policy, it should be transparent and ascertainable” (emphasis added). We note the qualification to this submission that it is only where there is or ought to be a policy that it should be transparent and ascertainable. However, it was also submitted that need for transparency was particularly acute in a post-conflict society. The authorities relied on by the appellant in support of these submissions were R (on the application of Salih and another) v Secretary of State for the Home Department [2003] All ER (D) 129 at paragraph [45]; Nadarajah v Secretary of State for the Home Department [2005] EWCA 1363 at paragraphs [67]–[68]; R (on the application of Justice for Health Limited) v Secretary of State for Justice [2016] EWHC 2338 at paragraph [141].
[28] Transparency as a constitutional and legal principle. The appellant’s submissions under this heading were connected to those under the previous heading. It was submitted that “transparency or open government as a constitutional principle developed by common law has now been clearly recognised.” Under this heading reference was made to the Parliamentary and Health Service Ombudsman who makes transparency and openness a core principle of good administration and to the Seven Principles of Public Life.
[29] The requirement for a policy in accordance with the Belfast Agreement and the NIA. The appellant relied on Re Rodger's Application in which consideration was given to the question as to whether there required to be a policy under the NIA in respect of the exercise of the Royal Prerogative of Mercy (“the RPM”). In that case it was stated at paragraph [86] that:
“a factor that indicates that there should be a policy are the references to the spirit of, but not the letter of, the Belfast Agreement or the spirit of, but not the letter of, the 1998 Act (see paragraph 16 of Terence McGeough’s Application for Judicial Review [2012] NICA 28). The spirit of an agreement or of an Act is a nebulous concept lacking definition particularly in relation to an agreement as one is searching for the spirit which has been agreed to by all those participating in the negotiations rather than in the subsequent and unilateral actions of one party to the agreement. The spirit of the Belfast Agreement or of the 1998 Act could mean different things to different people. If that was the defining feature in relation to the operation of the RPM then decisions could be made in an arbitrary manner and there would be a need for a policy not only for the benefit of the individual offenders but also to reassure the public as to the operation of such a significant part of the political settlement that occurred on 10 April 1998.”
It was submitted that just as the spirit of the Belfast Agreement was a nebulous concept lacking definition so also were the powers contained in section 1 and Schedule 1 NIA. This in turn led to the submission that a policy was required “to reassure the public as to the operation of such a significant part of the political settlement that occurred on 10 April 1998.”
[30] An implied obligation derived from the NIA to publish a policy. It was recognised by Mr Lavery that the Belfast Agreement and section 1 and Schedule 1 NIA contains no express duty to publish a policy. However, it was a feature of the appellant’s submissions that an implied obligation to publish a policy could be derived from the terms of the NIA taken with the provisions of the Belfast Agreement.
[31] Wednesbury unreasonable. Mr Lavery submitted that even if it was lawful for the respondent not to have a policy it would be Wednesbury unreasonable for him not to have one.
[32] The electorate. The submission on behalf of the appellant was that the first step in making an assessment under paragraph 2 of Schedule 1 NIA was to identify the electorate. It was said that before the respondent could assess whether a majority of those voting would favour a particular outcome it was essential for him to have decided who could vote. On this basis it was submitted that the respondent could not properly decide whether he is required to discharge his duty under paragraph 2 until he establishes the electorate by first choosing the criteria for the eligibility to vote. It was also submitted that the complexion of the electorate will have a significant bearing on the likely outcome of the border poll. In this way it was suggested that determining the electorate could determine whether a border poll should be directed under paragraph 2. For example, it was suggested that the following demographics may reasonably be enfranchised or, alternatively, disenfranchised: 16 years olds (as happened in the Scottish referendum); those born in Northern Ireland but ordinarily resident elsewhere; foreign or EU nationals who have been resident in Northern Ireland for a given period of time; foreign or EU nationals who are naturalised in Northern Ireland; prisoners; or anyone ordinarily resident in the Northern Ireland at any given time.
The first instance judgment
[33] The judge dismissed the challenge in a notably detailed and comprehensive judgment. His reasoning may be summarised as follows:
(i) There is no generally applicable common law requirement for public bodies to publish guidelines establishing how statutory powers will be exercised, the factors which will be taken into account or the sources of evidence.
(ii) No express or implied duty to publish a policy on when or whether to hold a border poll is found in the NIA. Reliance upon the Belfast Agreement is misplaced, as the statutory power is a reflection of what was agreed between the two governments and is contained in the British Irish Agreement in the Annex to the Belfast Agreement.
(iii) Where the exercise of statutory powers may interfere with Convention Rights and where the interference must be “in accordance with law”, the quality of law test may require guidance in order to avoid arbitrariness and to ensure that the law is sufficiently accessible and predictable.
(iv) The discretionary power to hold a border poll could be exercised lawfully for a broad range of reasons, irrespective of whether the duty to do so arose and even if it was believed that a majority might vote to remain part of the United Kingdom or if there was doubt about the issue. These could include giving a quietus to the issue for a period of time. The precise circumstances and context are variable and depend upon the exercise of complex political judgments which requires flexibility.
(v) The NIA does not specify the factors to be taken into account by the respondent prior to exercising the discretionary power to hold a border poll. It is therefore for the respondent to determine those factors in forming the necessary political judgment (applying Re Porter [2008] NIQB 10 at [51]).
(vi) The respondent’s decision not to publish a policy is a rational one. An attempt to pre-determine the factors to be taken into account or the evidence to be relied upon may prove unduly restrictive and not in the public interest. A policy worded in undefined and flexible terms would add nothing to the existing statutory wording and it was rational for the respondent not to adopt such a policy.
(vii) The duty upon the respondent to call a border poll necessarily implies that he will first reflect honestly upon the available evidence about whether a majority of the people of Northern Ireland would vote to form a united Ireland. It is for the respondent to determine the necessary evidence, which may include opinion polls and election results, but may also include a wide range of factors and sources, depending upon the prevailing circumstances at the time.
The Belfast Agreement
[34] At issue in this case are the provisions of section 1 and Schedule 1 NIA. The background to the enactment of those provisions is the Belfast Agreement.
[35] The aims of the Belfast Agreement included establishing a new devolved Government for Northern Ireland in which Unionists and Nationalists would share power. Its contents also included issues relating to sovereignty, civil and cultural rights, decommissioning of weapons, demilitarisation, justice and policing.
[36] The Belfast Agreement is made up of two inter-related documents.
[37] The first document is an “Agreement reached in the multi-party negotiations” (“the Multi-Party Agreement”). It is an agreement signed on behalf of the British and Irish Governments and eight political parties or groupings in Northern Ireland: the Ulster Unionist Party, the Social Democratic and Labour Party, Sinn Féin, the Alliance Party, the Progressive Unionist Party, the Northern Ireland Women's Coalition, the Ulster Democratic Party and Labour (“the political parties”). The Democratic Unionist Party, which has later became the largest unionist party, did not support the Belfast Agreement.
[38] In that section of the Multi-Party Agreement headed “Constitutional Issues” the participants (that is the two Governments and the political parties) endorsed the commitment made by the British and Irish Governments that, in a new British-Irish Agreement replacing the Anglo Irish Agreement, they will for instance:
(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.
[39] There are annexes to this section of the Multi-Party Agreement. Annex A is headed “Draft Clauses/Schedules for Incorporation in British Legislation.” Annex B is headed “Irish Government Draft Legislation to Amend the Constitution”. The draft Clauses/Schedules for incorporation in British legislation includes in Section 1(1) a declaration “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.” In Section 1(2) it includes the following “but if the wish expressed by a majority in such a poll is that Northern Ireland should cease to be part of the United Kingdom and form part of a united Ireland, the Secretary of State shall lay before Parliament such proposals to give effect to that wish as may be agreed between Her Majesty’s Government in the United Kingdom and the Government of Ireland.” Schedule 1 paragraph 1 provides that “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.” Schedule 1 in paragraph 2 provides that “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 shall cease to be part of the United Kingdom and form part of a United Ireland.”
[40] The second document which forms part of the Belfast Agreement was an “Agreement between the Government of the United Kingdom of Great Britain and Northern Ireland and the Government of Ireland” (“the British-Irish Agreement”). It was a draft international Treaty between two sovereign Governments which subsequently was executed by an exchange of diplomatic notes. It was annexed to the Multi-Party Agreement and the Multi-Party Agreement was annexed to it. In Article 2 of the British-Irish Agreement the two Governments affirm their solemn commitment to support, and where appropriate to implement, the provisions of the Multi-Party Agreement. Article 4 of the British-Irish Agreement provided that before it entered into force certain requirements had to be fulfilled including (a) British legislation shall have been enacted for the purpose of implementing the provisions of Annex A to the section entitled “Constitutional Issues” of the Multi‑Party Agreement; and (b) the amendments to the Constitution of Ireland set out in Annex B to the section entitled “Constitutional Issues” of the Multi-Party Agreement shall have been approved by Referendum. In the event those requirements having been fulfilled the British–Irish Agreement came into force on 2 December 1999.
[41] As part of the British-Irish Agreement, the British Government agreed to repeal the Government of Ireland Act 1920 which had established Northern Ireland, partitioned Ireland and asserted a territorial claim over all of Ireland. The Irish Government agreed to propose draft legislation to amend Articles 2 and 3 of the Constitution of Ireland, which asserted a territorial claim over Northern Ireland. Those conflicting territorial claims were to be removed and it was to be “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” (emphasis added).
[42] The Belfast Agreement was approved by voters across the island of Ireland in two referendums held on 22 May 1998. In Northern Ireland, voters were asked in the 1998 Northern Ireland Good Friday Agreement referendum whether they supported the multi-party agreement. In the Republic of Ireland, voters were asked whether they would allow the state to sign the agreement and allow necessary constitutional changes to facilitate it. There was overwhelming support in both jurisdictions: In Northern Ireland in support of the Agreement and in Republic of Ireland for both the Agreement and for the consequential Constitutional changes.
[43] As we have indicated in the British-Irish Agreement the British government agreed to enact the draft clauses/schedules in the Multi-Party Agreement dealing with the holding of a poll as to whether Northern Ireland shall cease to be a part of the United Kingdom and form part of united Ireland. Those draft clauses/schedules were incorporated into British legislation by section 1 and Schedule 1 of the NIA.
The NIA
[44] Before considering the appellant’s grounds of appeal we will set out and give consideration to the provisions of the NIA.
[45] The provisions of the Belfast Agreement which make reference to a border poll do not themselves have the force of law, JH Rayner (Mincing Lane) Ltd v Department of Trade and Industry [1990] 2 AC 418 at pages 476-477. They formed part of a political agreement and a draft international Treaty. As we have indicated they have the force of law, by means of section 1 and Schedule 1 NIA, which form part of the domestic law of the UK containing fundamental constitutional provisions for Northern Ireland.
(a) The interpretative approach to the NIA
[46] The question arises as to the court’s interpretative approach to section 1 and Schedule 1 of the NIA.
[47] The British–Irish Agreement, a Treaty is an aid to the interpretation of the NIA, see Bennion on Statutory Interpretation 7th Edition at 24.16 which states that “(when) a statute is passed in order to give effect to the United Kingdom’s obligations under a Treaty, the statute should if possible be given a meaning which conforms to that of the Treaty. For that purpose the provisions of the Treaty may be referred to as an aid to interpretation.” For the purposes of this appeal we also proceed on the basis that the Multi-Party Agreement is also an aid to the interpretation of the NIA, see Re McComb’s Application for Judicial Review [2003] NIQB 47 at paragraph [31].
[48] In JR80’s Application [2019] NICA 58 at paragraph [58] this court stated that:
“Lord Bingham in Robinson v Secretary of State for Northern Ireland and others [2002] NI 390 categorised the NIA as in effect a constitution whilst recognising that it did not set out all the constitutional provisions applicable to Northern Ireland. He continued at paragraph [11] by stating that “(so) to categorise the 1998 Act is not to relieve the courts of their duty to interpret the constitutional provisions in issue.” He added that “the provisions should, consistently with the language used, be interpreted generously and purposively, bearing in mind the values which the constitutional provisions are intended to embody.”
This court went on to observe that:
“It can be seen that the generous and purposeful interpretation has to bear in mind the values which the constitutional provisions are intended to embody.”
This court added that “(one) of those values is the democratic ideal but Lord Bingham identified other values which “this constitution is also seeking to promote.” One of the other values identified by Lord Bingham was the value of “participation by the unionist and nationalist communities in shared political institutions ….”
[49] In JR80 this court considered the value of participation by the unionist and nationalist communities in shared political institutions to be an extremely significant value. In that case this court recognised that there was a balance to be struck between government by civil servants on the one hand and on the other the reintroduction of direct rule which might persist and will adversely impact on the ability of unionist and nationalist communities to participate in shared political institutions. Anything that undermined that value was a legitimate matter to be taken into account in a generous and purposeful interpretation of the NIA.
[50] In this case we emphasise another aspect of Lord Bingham’s speech in Robinson. At paragraph [12] he stated that “where constitutional arrangements retain scope for the exercise of political judgment they permit a flexible response to differing and unpredictable events in a way which the application of strict rules would preclude.” That was one of the general considerations which Lord Bingham stated at paragraph [13] had a bearing on the statutory provisions in the NIA “at the heart of that case.” Lord Hoffman in his speech at paragraph [30] relied upon “the flexibility which could allow scope for political judgment in dealing with the deadlocks and crises which were bound to occur.” We consider that the exercise of the powers under section 1 and Schedule 1 NIA involve political judgment in the context of differing and unpredictable events. Accordingly, we consider that a flexible response in such circumstances is a consideration to be taken into account in a generous and purposeful interpretation of the constitutional provisions in section 1 and schedule 1 of the NIA. We consider that any policy in relation to directing the holding of a border poll which was not flexible would be inconsistent with that general consideration. Furthermore, we consider that if a policy was sufficiently flexible so as to be consistent with that general consideration it would be worthless merely repeating the obligation to form an assessment of the prevailing circumstances and the obligation to be honest and rigorously impartial in the context that it is for the people of the island of Ireland alone to exercise their right of self‑determination.
[51] We also consider that it would be legitimate for the respondent to consider the impact upon the value of participation by the unionist and nationalist communities in shared political institutions which could be caused by the process of consultation prior to and then the subsequent publication of a policy dealing with directions to holding a border poll. The respondent might form the view that such a process together with the publication of such a policy would be highly contentious disrupting the value of participation by the unionist and nationalist communities in shared political institutions. If that was so then in addition to contravention of the value of flexibility a policy would be contrary to the value of participation underpinning the constitutional arrangements for Northern Ireland set out in the NIA.
(b) Section 1 of the NIA
[52] Section 1 of the NIA under the rubric “Status of Northern Ireland” provides:
“(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.
(2) But if the wish expressed by a majority in such a poll is that Northern Ireland should cease to be part of the United Kingdom and form part of a united Ireland, the Secretary of State shall lay before Parliament such proposals to give effect to that wish as may be agreed between Her Majesty’s Government in the United Kingdom and the Government of Ireland.”
[53] There are a number of points to be made in relation to the terms of Section 1.
[54] The section is a word-perfect reproduction of the draft legislation contained in Annex A to that part of the Multi-Party Agreement dated 10 April 1998 headed “Constitutional Issues.” It is exactly what was agreed between the two governments and the political parties. It is also what was approved in referendums in both parts of the island of Ireland. It is consistent with the democratic ideal. In so far as Section 1 is concerned it fulfils the requirement in Article 4(1)(a) of the British-Irish Agreement for the entry into force of that Treaty.
[55] As a matter of domestic law Section 1 means that the status of Northern Ireland as a part of the United Kingdom is expressly recognised and will not change without the consent of a majority of its population voting in a poll convened in accordance with Schedule 1.
[56] The issue to be determined in that poll is whether Northern Ireland should remain part of the United Kingdom or whether it should cease to be part of the United Kingdom and form part of a united Ireland.
[57] The judge set out the inter-relationship between a border poll in both parts of the island of Ireland. He stated at paragraph [5]:
“It is clear that a border poll in Northern Ireland to produce the outcome of a united Ireland would have to be replicated by a poll in the Republic of Ireland producing a concurrent expression of a majority wish in the Republic to bring about a united Ireland. In effect if not de jure there would have to be an agreement between the UK and the Republic to have parallel polls in each jurisdiction. A vote in the north in a vacuum would not produce a united Ireland and in any event following majority votes north and south in favour of unification agreement would have to be reached between the UK and Ireland as to the form of that united Ireland and the way it which it would be governed and structured. The question arises whether any agreement between the two governments as to the nature of the united Ireland to follow from the votes would itself have to involve the consent of the majority in Northern Ireland. It would be likely to require changes in the Irish Constitution which would require the consent of the people in that jurisdiction.”
We agree that there is such an inter-relationship which must involve both governments. This inter-relationship led the judge to state that:
“The legal, practical and economic complications involved in unifying the country would be considerable. All this points to the conclusion that any decision as to the holding of a border poll involves extremely complex political considerations and if not carefully handled taking account of prevailing circumstances it could give rise to great instability” (emphasis added).
Again, we have no hesitation in agreeing with that conclusion about which there was no dispute at the hearing of this appeal. Rather on behalf of the appellant there was recognition of the complexity and the potential for instability but this it was said was all the more reason to deal with these issues calmly in advance by the respondent publishing a policy after appropriate consultation. To our minds that submission leaves out of account the judge’s reference to “prevailing circumstances.” It is obvious that circumstances have changed for instance with Brexit. A majority in Northern Ireland supported remain. It is equally obvious that circumstances will change. In turn changes in the prevailing circumstances will impact on and change the “extremely complex political considerations.” Thus the need for flexibility rather than consistency. The context does not dictate a lack of variation over time with a rigid approach but rather “a flexible response to differing and unpredictable events in a way which the application of strict rules would preclude .”
(c) Schedule 1 of the NIA
[58] Schedule 1 under the rubric “Polls for the purposes of section 1” provides:
“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.
4.(1) An order under this Schedule directing the holding of a poll shall specify—
(a) the persons entitled to vote; and
(b) the question or questions to be asked.
(2) An order—
(a) may include any other provision about the poll which the Secretary of State thinks expedient (including the creation of criminal offences); and
(b) may apply (with or without modification) any provision of, or made under, any enactment.”
[59] There are a number of points to be made in relation to the terms of Schedule 1.
[60] The wording of Schedule 1 is almost a word-perfect reproduction of the draft legislation contained in Annex A. Schedule 1 paragraphs 1-3 inclusive are word‑perfect reproductions of the draft legislation in Annex A.
[61] We say “almost” for two reasons.
[62] First paragraph 4 of Schedule 1 had not been drafted at the time of the Belfast Agreement. Rather paragraph 4 in Annex A stated:
“(Remaining paragraphs along the lines of paragraphs 2 and 3 of existing Schedule 1 to 1973 Act.)”
The “1973 Act” is the Northern Ireland Constitution Act 1973 section 1 of which made provision for a border poll to be directed in accordance with Schedule 1. Paragraphs 2 and 3 of Schedule 1 provided:
“2. Any order under this Schedule directing the holding of a poll shall make provision as to the persons entitled to vote on the poll, the question or questions to be asked of the persons so voting and the conduct of the poll, and may make such other provision in connection with the poll as appears to the Secretary of State to be expedient, including provision applying, with or without modifications, any enactment or statutory provision with respect to Parliamentary elections or elections to the Assembly.
3. The power to make orders under this Schedule includes power to vary or revoke a previous order and shall be exercisable by statutory instrument but no such order shall be made unless a draft of the order has been approved by resolution of each House of Parliament” (emphasis added).
The terms of Schedule 1 paragraph 4 NIA are very similar but not identical to paragraphs 2 and 3 of existing Schedule 1 to 1973 Act. We consider that the differences do not detract from the proposition that Schedule 1 is in effect a word-perfect reproduction of the draft legislation contained in Annex A.
[63] Secondly, the nature of the order to be made by the respondent was not specified in the draft of Schedule 1 in Annex A. Rather this was referred to as the remaining paragraphs were to be “along the lines of paragraphs 2 and 3 of existing Schedule 1 to 1973 Act.” As can be seen from the words to which we have added emphasis in paragraph 3 of Schedule 1 to 1973 Act the order was to be by way of a statutory instrument approved by resolution of each House of Parliament. An equivalent provision was set out, not in Schedule 1 NIA but rather in Section 96(2) which provides that an order under Schedule 1 “(a) shall be made by statutory instrument; and (b) shall not be made unless a draft has been laid before and approved by resolution of each House of Parliament.” Again we consider that the wording of Schedule 1 combined with Section 96(2) is effectively a word-perfect reproduction of the draft legislation contained in Annex A. We also consider that the Multi-Party Agreement and the British-Irish Agreement provided not only that the respondent had a role in the making of the order but so also did each of the Houses of Parliament. To our minds this emphasises the essentially political and democratic decisions to be made under paragraphs 1 and 2. Political in the sense that a decision having been made by the respondent, a politician it has to be positively endorsed by other politicians in both Houses of Parliament. Democratic in that the process of laying the draft before both Houses of Parliament ensures that the order is overseen by political representatives.
[64] We consider that Schedule 1 is in effect a word-perfect reproduction of the draft legislation contained in Annex A to that part of the Multi-Party Agreement dated 10 April 1998 headed “Constitutional Issues.” It is exactly what was agreed between the two governments and the political parties. It is also what was approved in referendums in both parts of the island of Ireland. It is consistent with the democratic ideal. In so far as Schedule 1 is concerned it fulfils the requirement in Article 4(1)(a) of the British-Irish Agreement for the entry into force of that Treaty.
[65] Paragraph 1 of Schedule 1 confers upon the respondent a discretion to direct the holding of a border poll whilst paragraph 2 imposes upon the respondent a duty to direct the holding of a border poll.
[66] Paragraph 3 of Schedule 1 prohibits the respondent from directing the holding of a border poll earlier than 7 years after the holding of a previous border poll. A similar provision was contained in paragraph 1 Schedule 1 to the 1973 Act though the period of time in that Act was “earlier than 10 years after a previous” border poll. There are four points in relation to paragraph 3 of Schedule 1 NIA. First, this prohibition applies to both a discretionary border poll under paragraph 1 and to the requirement to hold a border poll under paragraph 2. So even if it appears likely to the respondent that a majority of those voting would express a wish to form part of a united Ireland a border poll cannot be held earlier than 7 years after the holding of a previous border poll. Second, paragraph 3 adds to the political nature of the decision to be made by the respondent under paragraph 1 as he may take into account as part of the prevailing circumstances the likelihood of a change in the wishes of the majority requiring the holding of a border poll under paragraph 2. Third, the respondent would not be acting with rigorous impartially if in the face of diminishing support for Northern Ireland remaining in the United Kingdom he directed the holding of a border poll with the sole purpose of achieving a majority to remain and thereby to delay a united Ireland for a period of 7 years. Fourth the terms of paragraph 3 as with the rest of the section 1 and Schedule 1 NIA were agreed between both governments and the political parties. The terms were also contained in the British-Irish Agreement and there was overwhelming support for the Belfast Agreement in referendums in Northern Ireland and in the Republic of Ireland. They are consistent with the democratic ideal.
[67] In summary paragraph 4 of Schedule 1 provides that the order and therefore the statutory instrument must specify the questions to be asked and the persons entitled to vote and may include any other provision about the border poll which the respondent considers to be expedient. There is no requirement in paragraph 4(1) to formulate the question in any particular way nor are there any suggestions as to who is entitled to vote. Whilst these matters must be specified in the order they are left to the respondent subject to the statutory instrument being approved by resolution of each House of Parliament. Paragraph 4(2) is extremely wide for instance permitting any other provision which the respondent thinks expedient. We consider that these powers in paragraph 4 must be exercised honestly in the public interest with rigorous impartiality in the context that it is for the people of Ireland alone to exercise their right of self-determination.
(i) Discretion to direct the holding of a border poll under paragraph 1
[68] The discretion to direct the holding of a border poll is unqualified. Schedule 1 paragraph 1 does not specify any matter which should be taken into account or any matter which should be left out of account.
[69] The exercise of discretion must be preceded by the respondent’s assessment of the prevailing circumstances. As we have indicated it is obvious that those circumstances have and will change over time.
[70] The discretion must be exercised honestly.
[71] The exercise of the discretion is based upon the respondent’s assessment of whether directing the holding of a border poll is in the public interest which assessment involves political judgment.
[72] An aid to the interpretation of what is in the public interest are the terms of the British-Irish Agreement, the international Treaty which was enacted in the NIA. Article 1 paragraph (ii) of the British-Irish Agreement provides recognition “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, …” (emphasis added). Also Article 1 paragraph (v) provides an affirmation that “…the power of the sovereign government with jurisdiction (in Northern Ireland) shall be exercised with rigorous impartiality on behalf of all the people in the diversity of their identities and traditions and shall be founded on the principles of full respect for, and equality of, civil, political, social and cultural rights, of freedom from discrimination for all citizens, and of parity of esteem and of just and equal treatment for the identity, ethos, and aspirations of both communities” (emphasis added). There was no issue that these two provisions meant that not only must the respondent act honestly in the exercise of discretion to direct a border poll but he must also act with rigorous impartiality in the context that it is for the people of the island of Ireland alone to exercise their right of self-determination.
[73] Schedule 1 contains no express duty to publish a policy as to when or in what circumstances it is in the public interest to hold a border poll.
(ii) Duty to direct the holding of a border poll under paragraph 2
[74] The duty to direct the holding of a border poll depends upon whether it appears likely to the respondent 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.
[75] The duty is triggered by the respondent’s assessment.
[76] The assessments under paragraphs 1 and 2 are different. Paragraph 1 involves an assessment of public interest. Paragraph 2 involves an assessment as to the likely majority of those voting.
[77] The duty under paragraph 2 arises even if it is not in the public interest to direct the holding of a border poll.
[78] Paragraph 2 does not specify any matter which should be taken into account or any matter which should be left out of account in the assessment.
[79] Paragraph 2 is silent as to the sources of information which the respondent might rely upon.
[80] The assessment involves an evaluative judgment as to a likely outcome. We consider that it is essentially a political judgment. It is assigned to and is to be performed by the respondent, a politician who is to form an assessment as to the political views of others. The political judgment as to the likely outcome of a border poll is not a simple empirical judgment driven solely by opinion poll evidence. It is also not a simple judgment based purely on perceived religion. The judgment depends on what are the prevailing circumstances at any given time. For instance a likely outcome may involve an evaluation as to whether there are other factors which will impact on voting intentions crossing traditional party or perceived religious lines and if so as to their impact. Instances of such factors are changes in social attitudes North and South, relative economic prosperity North and South, the taxation structures North and South, the outcome of Brexit and the nature of future trading relations between both parts of Ireland which in turn depends on any agreement between the United Kingdom and the European Union.
[81] The NIA contains no express duty to publish a policy as to how the respondent should assess whether there is an obligation to direct the holding of a border poll.
[82] The judge stated that it “is necessarily implied in (Schedule 1 paragraph 2) that the Secretary of State must honestly reflect on the evidence available to her to see whether it leads her to the conclusion that the majority would be likely to vote in favour of a united Ireland.” We agree that the respondent must act honestly. We would add that he must also act with rigorous impartiality in the context that it is for the people of the island of Ireland alone to exercise their right of self-determination.
The grounds of appeal
[83] We have summarised the submissions relied on by the appellant which in turn reflect the grounds of appeal. Before we deal individually with each of the submissions we conclude on an overall basis that all of them contravene the constitutional value of flexibility which we have set out at [50] and could also undermine the value of participation in the manner set out at [51]. We reject all of the appellant’s submissions.
(a) Inconsistency
[84] The appellant submits that the duty to introduce a policy derives from the principle that “inconsistency is a ground for judicial review.” The appellant suggests that an obligation to publish a policy flows from a general principle of acting consistently.
[85] First, we reject the proposition that the powers contained in section 1 and Schedule 1 require consistency. Rather as we have explained the exercise of those powers involve political judgment in the context of differing and unpredictable events. Accordingly, we consider that a flexible response is required in accordance with the interpretative approach to the NIA as explained in Robinson. Even if inconsistency is a ground for judicial review it has no application to the circumstances of this case.
[86] Second, we reject the proposition that “inconsistency is a ground of judicial review” except in that it is evidence of Wednesbury irrationality. The question as to whether consistency is a ground of judicial review was considered and rejected by the Supreme Court in R (Gallagher Group & Ors) v Competition and Markets Authority [2018] UKSC 25. It appears from that authority that consistency is a “generally desirable” objective but not an absolute rule. Lord Carnwarth stated at paragraph [24]:
“Whatever the position in European law or under other constitutions or jurisdictions, the domestic law of this country does not recognise equal treatment as a distinct principle of administrative law. Consistency, as Lord Bingham said in the passage relied on by the appellant …, is a “generally desirable” objective, but not an absolute rule.”
In this way consistency is a component of equal treatment which in turn is not a separate principle, but part of Wednesbury rationality. It has arisen in cases where it was argued that administrative decision makers had treated different individuals or classes of individual differently in arguably comparable circumstances. For instance in Alconbury, the House of Lords recognised that publication of a policy was a “perfectly proper course, for the provision of guidance in the exercise of administrative discretion.” Alconbury was a planning case, which is classically an area where consistency is desirable and where guidance is important not only for planning authorities, but developers when making investment decisions. The nature of the decisions in Alconbury are entirely different from the nature of the decisions to be made under Section 1 and Schedule 1 NIA. The respondent’s discretion under paragraph 1 of Schedule 1 and his duty under paragraph 2 of Schedule 1 NIA do not require him to act in an adjudicative or regulatory capacity. They involve him making political judgments about whether it is in the public interest to hold a border poll and as to whether it appears likely to him that a majority of those voting would express a wish to form part of a united Ireland. These political judgments do not involve analysis of “comparator” cases with which the respondent might otherwise be required to act consistently.
[87] We also note that “consistency” as a ground of judicial review had previously been considered in this jurisdiction in Re Croft [1997] NI 457 and in Re Morrison [1998] NI 68. In Morrison Kerr J held that where a public authority had published a policy explaining how its discretion would be exercised, it was “normally” required to act consistently with the policy but that no such obligation arose where there was no policy. The requirement to act in a consistent manner (in the absence of justification) flowed from the existence of a published policy not as the appellant contends in this case that an obligation to publish a policy arises from a general principle to act consistently.
(b) Policy and the rule of law
[88] As we have indicated the appellant relies upon a number of authorities to support the proposition that a failure to publish a policy would contravene the rule of law. We do not consider that those authorities support such a broad principle. Rather they support two separate principles, namely:
(i) Where ECHR rights are engaged, the requirement that an interference “is in accordance with the law” may require publication of a policy. In Malone v United Kingdom (1985) 7 EHRR 14 at paragraphs [66] - [68] the ECtHR considered the general principles governing whether the interference found in that case was “in accordance with the law.” Those principles included that the interference in question must have some basis in domestic law; that there must be compliance with the domestic law; that the law must be adequately accessible; and that a norm cannot be regarded as “law” unless it is formulated with sufficient precision to enable the citizen to regulate his conduct.
(ii) As a matter of domestic administrative law, where a public authority has formulated and applies a policy, it should be published.
[89] The first principle is of no relevance to this appeal, since the appellant in his Order 53 statement has not relied on any ECHR rights.
[90] The second principle does not arise in this appeal as there is no policy and therefore there is nothing to be published.
[91] We do not consider it necessary to analyse all the cases relied on by the appellant but rather we will illustrate by reference to B v Secretary of State for Work and Pensions. The appellant relied upon a quotation from the judgment of Sedley LJ at paragraph [43]. The part of paragraph [43] relied on by the appellant was as follows:
“It is axiomatic in modern government that a lawful policy is necessary if an executive discretion of the significance of the one now under consideration is to be exercised, as public law requires it to be exercised, consistently from case to case but adaptably to the facts of individual cases.”
However, paragraph [43] continued in the following terms:
“If - as seems to be the situation here - such a policy has been formulated and is regularly used by officials, it is the antithesis of good government to keep it in a departmental drawer. Among its first recipients (indeed, among the prior consultees, I would have thought) should be bodies such as the Child Poverty Action Group and the Citizens Advice Bureaux. Their clients are fully as entitled as departmental officials to know the terms of the policy on recovery of overpayments, so that they can either claim to be within it or put forward reasons for disapplying it, and so that the conformity of the policy and its application with principles of public law can be appraised, although two such policies were evidently described or shown to Newman J in R (on the application of Larusai) v Secretary of State for Work and Pensions [2003] EWHC 371 Admin: see para 15 and 19.”
That case concerned the Secretary of State’s discretion to recover benefits which had been overpaid. The Court of Appeal decided that the Secretary of State had the power to make a recovery decision, but also a discretion whether to do so. Sedley LJ commented upon a submission to the effect that the Secretary of State had a policy on such decisions, which had not been published. We consider that the comments at paragraph [43] related to the need to publish a policy which already existed. We also consider that the first part of paragraph [43] is not a broad statement of constitutional principle defining an obligation to publish a policy. Rather it is recognition that policies can be of great importance to officials exercising broadly expressed discretionary powers when making administrative adjudications. Those administrative adjudications are completely different from the exercise of the powers in section 1 and Schedule 1 NIA which involves political judgement in the context of differing and unpredictable events.
[92] The appellant also relied upon my decision in Re Rogers which concerned the exercise of the Royal Prerogative of Mercy. The applicant in that case contended that there was a requirement to have a policy governing the exercise of RPM for persons sentenced after the Belfast Agreement who were subject to the accelerated release provisions of section 10 of the Northern Ireland (Sentences) Act 1998. The appellant relied upon dicta at paragraph [86] in which I referred to the need for a policy in order to avoid arbitrariness in release decisions, both for the benefit of prisoners and the public. The respondent submits that the comments at paragraph [86] were directed towards the desirability of a published policy, rather than a legal obligation to do so given my conclusion that a policy was not required. We agree. Furthermore, at paragraph [87] I stated:
“I consider that the operation of the RPM in the context of section 10(6) of the 1998 Act is “likely to be highly dependent on the particular facts of each case, facts which will almost certainly vary greatly from one case to another” (see paragraph 14 of McGeough). I also consider that no policy should undermine the legislative intent that those subsequently convicted should serve two years imprisonment before being entitled to accelerated release. On those grounds I do not consider that this is an area which is amenable to a policy which could conceivably cover the factual situations which might arise. Any policy that was created could only reiterate the legislative intent that a person subsequently convicted should serve two years in relation to any sentence imposed before being entitled to accelerated release and then go on to state that each case will be considered on its particular facts. I consider that the number of occasions upon which decisions require to be made are not so numerous that a policy is necessary to ensure consistency from case to case. The RPM has been exercised 16 times in the last 14 years. It has not been exercised in Northern Ireland since 2002. I am also satisfied that it was perfectly possible to bring all the facts in relation to the applicant’s case to the attention of the Secretary of State without there being a policy in existence” (emphasis added).
We consider that reasoning at paragraph [87] can be read across to the facts of this appeal. The statutory discretions in Section 1 and Schedule 1 can be subject to policy decisions at any given time in the context of “dealing with deadlocks and crises which are bound to occur” but they are not amenable to an enduring policy which would bind the respondent now and in the future as to how the flexible and politically sensitive powers are to be exercised. In that sense this also is an area which is not amenable to a policy. Indeed, such a policy would not accord with the requirement for flexibility as explained in Robinson. Section 1 and Schedule 1 is another example of statutory discretions in which such a policy is not appropriate.
(c) Transparency and the requirement for an adequate policy
[93] The appellant relies upon paragraph [55] of the judgment of Lord Mance in Kennedy v Charity Commission in support of the proposition that “transparency” is a common law constitutional principle. Lord Mance stated:
“……But the right approach is now surely to recognise, as De Smith's Judicial Review (7th edn, 2013) para 11–028 suggests, that it is inappropriate to treat all cases of judicial review together under a general but vague principle of reasonableness, and preferable to look for the underlying tenet or principle which indicates the basis on which the court should approach any administrative law challenge in a particular situation. Among the categories of situation identified in De Smith are those where a common law right or constitutional principle is in issue. In the present case, the issue concerns the principles of accountability and transparency, which are contained in the 1993 Act and reinforced by common law considerations and which have particular relevance in relation to a report by which the Charity Commission makes to explain to the public its conduct and the outcome of an inquiry undertaken in the public interest” (emphasis added).
In Kennedy the Charity Commission had the general statutory duty to have regard to principles of accountability and transparency. The words to which we had added emphasis are to be seen in the context of the Charities Act which enshrines transparency. We do not consider that Kennedy established a generally applicable constitutional principle of transparency. We reject that there is a constitutional principle of transparency that requires the publication of a policy in relation to section 1 and Schedule 1 NIA.
(d) Transparency as a constitutional and legal principle
[94] As we have indicated the appellant’s submissions under this heading were connected to those under the previous heading. It was submitted that “transparency or open government as a constitutional principle developed by common law has now been clearly recognised.” We do not consider that this proposition, even if correct could lead to the requirement to publish a policy particularly given the constitutional value of flexibility which we have set out at [50] given that the exercise of the powers under section 1 and Schedule 1 NIA involve political judgment in the context of differing and unpredictable events.
(e) The requirement for a policy in accordance with the Belfast Agreement and the NIA
[95] As we have indicated there is no requirement for a policy in either the Belfast Agreement or in the NIA. Rather as we have explained the exercise of those powers involve political judgment in the context of differing and unpredictable events. Accordingly, we consider that a flexible response is required in accordance with the interpretative approach to the NIA as explained in Robinson.
(f) An implied obligation derived from the NIA to publish a policy
[96] We consider that an obligation to publish a policy cannot be implied as it is not necessary in order to give effect to the Multi-Party Agreement, the British-Irish Agreement or to the NIA. We consider that the implication of such an obligation would be inconsistent with the interpretative approach to the NIA since it could constrain the exercise of powers which were intended by both Governments, the political parties and by Parliament to be broad and sufficiently flexible to accommodate the changing and uncertain circumstances which might arise in Northern Ireland in the future.
(g) Wednesbury unreasonable
[97] The suggestion that the respondent’s decision not to have a policy was Wednesbury unreasonable was only faintly made. We reject that suggestion. The rationality of the decision is clear given what was agreed in the Multi-Party Agreement and in the British-Irish Agreement. Those agreements did not specify that a policy was required rather two sovereign governments and the political parties invested the power in the respondent without any constraints. Furthermore, the decision not to have a policy is rational given the constitutional values which we have set out at [48]–[51] in the context that the exercise of the powers under section 1 and Schedule 1 NIA involve political judgment in the context of differing and unpredictable events.
[98] The provisions of section 1 and Schedule 1 NIA do not specify any particular matters which must, as a matter of duty, be taken into or left out of account by the respondent in deciding on whether to direct a border poll. In essence it is for the respondent’s judgment to decide which matters it should take into or leave out of account subject to Wednesbury considerations. On that basis we consider that it is for the respondent to decide what is, or is not relevant to the decision-making process depending on the prevailing circumstances.
(h) The electorate
[99] We accept that there is an inter-relationship between the decision as to who is entitled to vote and an assessment as to the likely outcome of that vote. In that respect the respondent is entitled to consider who should vote whenever making a decision as to the likely outcome of a border poll. This was the conclusion of the judge in paragraph [21] in which he stated that:
“On the question of determining who should vote in a border poll that is a question that falls to be determined when the Secretary of State concludes that a poll should be ordered. In deciding how she thinks the majority would vote in a poll under Schedule 1 paragraph 2 the Secretary of State is entitled to consider what she considers would be the likely pool of voters that pool being the one to be chosen by the Secretary of State in the exercise of powers under Schedule 1 paragraph 4. She is not required as a matter of law to enunciate a policy on how the pool of voters should be determined in advance of her exercising her powers under the Schedule.”
We agree.
[100] We add that a decision as to who should vote is also a political judgment as to what is acceptable or appropriate in our community. That involves political judgment in the context of differing and unpredictable events. An instance is in relation to lowering the voting age to 16 as in the Scottish referendum. We accept that this may have a considerable impact on the outcome of a border poll. The present voting age in Northern Ireland is 18. However, there may be a changing consensus in the island of Ireland as to the appropriate voting age which would form a component of determining what voting age was in the public interest in relation to a border poll. We consider that the judgment formed in Northern Ireland may or may not be the same as in the Republic of Ireland but that whatever decision is made in the Republic of Ireland might be a factor to be taken into account by the respondent in Northern Ireland. We consider that the constitutional value of flexibility requires to be maintained and that it would not be maintained by the publication of a policy as to present views in relation to voting age or in relation to any other present decision as to demographics.
Conclusion
[101] We dismiss the appeal for the reasons which we have given.
[102] We will hear counsel in relation to costs.