Introduction – a perfect storm?

As predicted, Brexit is proving to be profoundly destabilising for the peace process and the constitutional politics of Northern Ireland. An outcome that lacks the consent of the people of Northern Ireland (a majority voted to remain) is re-opening fundamental questions about future relationships across these islands. We argue that this constitutional mess has potentially created a ‘perfect storm’, and leaves many here struggling with the troubling consequences.

The March 2017 White Paper sets out plans for the Great Repeal Bill (GRB) to incorporate, through conversion or transposition, most EU law into domestic legislation, including laws in devolved areas of competence. These policy areas are more extensive in Northern Ireland than Scotland or Wales, and include employment, social security and equality law. Further to the Good Friday Agreement (GFA), and other accords making up the peace settlement, the constitutional arrangements for unionist-nationalist power-sharing also differ to the other devolution settlements. They encompass an array of mutual vetoes for parties who take diametrically-opposed views on Brexit, the virtues of the GFA arrangements and on issues of rights protection, including those that have stemmed from EU law.

If the devolved institutions are re-established (as a result of ongoing negotiations) the GRB will become particularly relevant to their potential stability. A scenario may emerge where the UK Government will either choose to allow its agenda to be radically checked by devolved power, or alternatively continue to disrespect the constitutional arrangements flowing from the GFA, with significant political consequences.

The fragile ‘trust us’ foundation of provisions in the peace settlement have already been shaken by the manner of triggering Article 50, where the Supreme Court held that devolved legislative consent provisions may well be constitutional conventions but are not subject to legal enforcement by the courts. This episode laid bare the limitations of conventions as a matter of domestic law, however significant they may be as components of political constitutionalism. It also signalled that the UK Government is willing to contest their applicability or depart from them in the context of advancing Brexit.

The initial hurdle for the GRB bill is the requirement for a Legislative Consent Motion (LCM) from the devolved administrations, a matter which, despite the bill clearly engaging the devolved matters that trigger an LCM, the White Paper is curiously silent on. There are also significant implications flowing from the proposed Ministerial powers to make subordinate legislation to amend or repeal primary legislation. The White Paper clarifies these ‘Henry VIII’ powers would extend to devolved Ministers, granting power to ‘amend devolved legislation to correct law that will no longer operate appropriately, in line with the power we propose should be held by UK ministers’.  This post further explores the implications of these matters and the perfect storm they are likely to create.

The Arrangements for Power-sharing and the Divergent Views of the Parties

The GFA gave life to the current power-sharing arrangements, underpinned with a bilateral UK-Irish treaty, and endorsed in concurrent referendums in the North and South of Ireland. Its provisions were largely implemented through the Northern Ireland Act 1998 (NIA); further amended in light of changes derived from the 2006 (UK-Ireland) St Andrews Agreement. There are a range of other safeguards provided for in the Agreements, intended to reinforce power-sharing, and that are yet to be implemented, not least a Bill of Rights for Northern Ireland.

In terms of legislative competence, the Northern Ireland Assembly is empowered to legislate (in a manner compatible with inter alia the ‘Convention rights’ and EU law) on all transferred matters. Such matters are anything that is not listed in Schedules to the Act setting out excepted (e.g. immigration, international relations) and reserved (e.g. civil aviation) matters. Notably the reference to international relations includes ‘relations with the EU’ but explicitly not observing and implementing obligations under EU law, which is explicitly a devolved competence (NIA 1998, schedule 2, para 3(c)). All members of the Assembly (MLAs) are to designate as ‘unionist’ ‘nationalist’ or ‘other’. 30 MLAs can lodge a ‘Petition of Concern’, which would then require sufficient support from both designated unionists and nationalists to ensure a division carries. Whilst envisaged as a ‘minority rights’ protection mechanism there are no criteria to lodge a Petition beyond the need for 30 MLAs, and conversely it has been used for the opposite purpose (including recently by the Democratic Unionist Party (DUP) to veto a motion on same-sex marriage equality). Under the 2015 Fresh Start Agreement a voluntary Protocol was agreed to limit the use of Petitions of Concern to ‘exceptional circumstances’. Debate continues on the future of this constitutional device.  It remains to be seen however whether its application will affect the use of petitions in relation to the significant elements of Brexit.

The office of the First and deputy First Minister is a joint office requiring consensus among both Ministers to act (now called The Executive Office – TEO). Since 2007 this office has been held by the DUP (designated unionist) and Sinn Féin (designated nationalist), as the first and second largest parties respectively. All parties with proportionately sufficient MLAs are entitled to Ministers within the power-sharing Executive. For most of the time since 2007 the Ulster Unionist Party (UUP – ‘unionist’), the SDLP (‘nationalist’) and Alliance (‘other’) have also participated in the Executive.  In the short- lived mandate from 2015 the only Executive parties were the DUP and Sinn Féin. That mandate collapsed following a political scandal over overspending on the ‘Renewable Heat Incentive’ scheme (RHI), among other things. Fresh elections in March 2016 were prompted by the resignation of the deputy First Minister, the late Martin McGuinness MLA (in part informed by the Speaker’s decision to allow the First Minister Arlene Foster MLA to make a unilateral statement to the Assembly on RHI without the approval of her deputy, thus bypassing the joint nature of the office). Sinn Féin has taken the position that they will not nominate a deputy First Minister, required to re-establish the institutions, unless safeguards from previous Agreements are implemented and ‘genuine’ power-sharing is instigated. It is this that is the principal focus of ongoing talks, currently paused for the UK General Election.

Under changes introduced through the St Andrews Agreement the power of Northern Ireland Ministers is qualified insofar as any ‘significant and controversial’ matter not in the Programme for Government must be agreed by the Northern Ireland Executive.  The Courts have already held a Ministerial decision on a John Lewis store ultra vires as it was sufficiently ‘significant and controversial’ to require full executive approval. It is reasonable therefore to read this provision as preventing any Minister acting unilaterally on any politically contentious matter.  The context of Brexit and the GRB provides for plenty of ‘significant and controversial’ matters:

• The DUP strongly supported the ‘Leave’ campaign, the other parties advocated for ‘Remain’;
• The DUP still oppose the Good Friday Agreement (a point too often skipped over), which the other parties support (this is particularly relevant to the aspects of the GFA relating to Ireland and Irish citizens);
• The DUP and UUP have taken the position (‘the unionist position’ – also advocated by the UK Government) that the referendum result should be considered on the basis of the UK vote, and thus provides a mandate for Northern Ireland to leave the EU;
• Sinn Féin and the SDLP have taken the position, (‘the nationalist position’ – also advocated by the SNP), that the constitutional arrangements require consideration of the result at jurisdictional level, and make the case that on the basis of the Northern Ireland vote (56% ‘remain’) there is no mandate to take Northern Ireland out of the EU;
• The Alliance party would like the decision to leave the EU ‘reconsidered’, but if this does not happen advocate for the softest Brexit possible.

The DUP also disagrees on aspects of equality law, including (in the recent mandate) blocking legislation to protect children from discrimination in relation to goods, facilities and services.  Whilst the DUP stands almost alone on some of the above issues, it remains the largest political party in Northern Ireland. In previous mandates the DUP had sufficient MLAs to invoke a ‘petition of concern’ and veto Assembly business on its own (this position changed following the Assembly election in March).

Whilst the DUP is strongly ‘Eurosceptic’, with its MEPs not sitting in any EP grouping, this does not necessarily mean that the party is averse to some elements of a ‘softer’ Brexit. The DUP has opposed, for example (in common with the other parties) a ‘hard border’ on the island of Ireland, and has recognised that there are special circumstances in Northern Ireland. However, it is evident that strongly divergent policy positions on Brexit still exist.

If the institutions return this will test their capacity and durability. It is hard to imagine a political project that could be more unhelpful to fostering stability in Northern Ireland. That is why Brexit has caused such dismay for all those genuinely concerned for long-term and meaningful peace and stability. Whether political parties are able to find common ground on the detailed requirements of special status/circumstances remains to be seen.

Legislative Consent, Brexit and the Great Repeal Bill

The GFA provides generally that the power of the Westminster Parliament to make legislation for Northern Ireland ‘will remain unaffected’ but then specifies that Westminster will legislate on non-devolved issues and also ‘legislate as necessary to ensure the United Kingdom’s international obligations are met in respect of Northern Ireland.’ (Paragraph 33 of Strand One)

S5(6) of the NIA (dealing with Acts of the Northern Ireland Assembly) is limited to reiterating that the UK Parliament retains power to make laws for Northern Ireland. It is the Devolution MoU  that sets out further particulars:

However, the UK Government will proceed in accordance with the convention that the UK Parliament would not normally legislate with regard to devolved matters except with the agreement of the devolved legislature. The devolved administrations will be responsible for seeking such agreement as may be required for this purpose on an approach from the UK Government.

There are therefore a number of issues in relation to legislative consent. First, the scope of the qualification that Westminster will not ‘normally’ legislate as regards devolved matters; second, the determination of what constitutes ‘with regard to devolved matters’, and third, the process that will be followed in order to seek legislative consent.

Whilst the qualification that Parliament will not ‘normally’ legislate on devolved matters is not elaborated on, it is notable that the GFA does set out one express circumstance:  that Parliament will legislate when it is necessary to meet the UK’s international obligations. There is one known precedent for such a situation – and it relates to an EU directive, namely the transposition of the Gender Directive (2004/13/EC) in 2007. The then deputy First Minister Martin McGuinness MLA, was willing to legislate but his First Minister partner in the joint office, the late Ian Paisley MLA, was ‘not agreeable’ reportedly due to the ‘explicit inclusion of reference to transgender or gender reassignment in the regulations’, in this circumstance UK Ministers intervened for the matter to be legislated for in Westminster.

This power is permissive; there are a broad range of unmet international obligations. This includes those flowing from the peace Agreements that have not been legislated for. For example, the Irish Language Act (committed to by the British Government in the bilateral UK-Ireland St Andrews Agreement) where Whitehall has steadfastly argued that as the Act would deal with a transferred matter there can be no legislation without ‘consensus’ in the Assembly. The UK Government leaves itself open to charges of inconsistency if it then departs from such a position on other matters, such as Brexit (where it is responding to a UK-wide vote as opposed to any international obligations).

It would be odd to suggest that the ‘normally’ provision is tailored to allow the UK Government not to seek legislative consent if it thinks the devolved institution will not agree. However, Government sought to studiously avoid any question of legislative consent in relation to the European Union (Notification of Withdrawal) Act 2017, in the knowledge that both the Scottish Parliament and Northern Ireland Assembly would vote an LCM down. Whilst this in itself would not have prevented Westminster from legislating, to do so on the basis of such express opposition would have had profound political implications in both jurisdictions. There are clear political costs to bypassing the convention.

There is room for interpretative disagreement over what constitutes a devolved matter, and this may become a ground of considerable political contention. A further question is as regards the process for an LCM, which is currently provided for in the Standing Orders of the Northern Ireland Assembly. The usual practice is for the relevant Minister to lay a legislative consent memorandum (inclusive of a draft LCM) before the Assembly, then following Committee scrutiny the competent Minister seeks Assembly approval. As noted, the Minister may require the approval of the full Executive to do this. This means the larger parties (the DUP and Sinn Féin) could reject the LCM themselves at this stage.

The UK Government has been willing to depart fairly dramatically from this. For example, in relation to the Crime and Courts Bill that set up the National Crime Agency (NCA). The original intention of the Home Office was for the NCA to operate in Northern Ireland, with policing powers but without being accountable to the relevant Northern Ireland mechanisms. This caused concern among human rights groups and, as policing powers were devolved, required an LCM that both nationalist parties (the SDLP and Sinn Féin) were not willing to support. Ultimately the Home Office shifted its position to a sufficient extent to convince the SDLP, but not Sinn Féin. In this context a new process was ‘invented’. A private members’ motion supporting the NCA legislation was tabled in the Assembly and passed a straight vote. The Speaker of the Assembly in response to a Point of Order from Sinn Féin clarified that the vote did not constitute an LCM. The UK Government however then accepted that this constituted the Assembly ‘consenting’ to the relevant precision of the Crime and Courts Bill and proceeded to legislate.

A further complication is that the general practice has been to seek an LCM for the relevant transferred provisions in an entire bill. This brings first the question of timing (an LCM at an early stage risks being a ‘blank cheque’ from the Assembly; Parliament could subsequently amend the provisions in question; at a late stage it gives little room for Parliament to consider an alternative approach if the LCM does not pass). Second, it is also possible – in theory at least – for different provisions of the bill to be subject to separate LCMs, which may be attractive to government if it believes certain provisions are more likely to meet resistance.  However, the Standing Orders are not predicated on a request from Westminster before a devolved Minister tables an LCM, in principle Stormont Ministers can table an LCM without such a request.

It is this context that the Great Repeal Bill falls into. GRB will repeal the ECA 1972, convert EU law into domestic law and will vest the new ‘Henry VIII’ powers in Ministers, including devolved ministers. The White Paper is explicit that the GRB will deal with transferred matters and make changes to the powers of devolved Ministers and the devolved legislature. Under the Assembly’s Standing Orders any one of these devolved matters on its own triggers the need for an LCM. All three are present within the GRB – all the more notable therefore that the White Paper makes no reference anywhere to the need for legislative consent.

There are a range of scenarios with an LCM. It is possible that parties taking the ‘nationalist position’ will block an LCM – for symbolic and/or substantive reasons, on grounds that it is part of the implementation of Brexit. The nationalist parties alone can table a Petition of Concern that would ensure an LCM would not pass, even if unionists and others supported it. Conversely those taking ‘the unionist position’ are unlikely to want to be seen to be standing in the way of Brexit implementation, but the DUP in particular may be opposed to some of the EU-originated provisions that the GRB incorporates, particularly those in relation to rights protection. The DUP may well be motivated to block some such positions, particularly if pressed by their unionist critics such as the TUV. For example, if this involved agreeing to incorporate EU anti-discrimination provisions on grounds of sexual orientation. The DUP no longer has the numbers to trigger a Petition of Concern on its own, and would be reliant on the assistance of other MLAs to do so.

As suggested here, the White Paper is shy of detail in its brief section on ‘interaction with the devolution settlements’. This section does set out that the GRB will contain provisions to ensure the ‘effective functioning of the UK single market is maintained’, to for example, ‘protect the freedom of businesses to operate across the UK single market and to enable the UK to strike free trade deals with third countries’ (White Paper, para 4.3). Whilst these references are vague, their appearance suggests an intention to qualify devolved powers to the extent they might interfere with a ‘UK internal market’. This could mean limiting Stormont’s powers on employment rights, procurement or other regulations on business; this issue itself could be sufficiently controversial that enough parties do not support an LCM. Whatever view is taken on this, it is possible that devolved consent is likely to be viewed as an irritant to the ambitions of the UK Government as it charts a path through Brexit.

The UK Government could opt to ignore the convention entirely or disregard an LCM vote and legislate regardless. Both options would have significant political consequences, essentially undermining the GFA and broader basis for the devolution settlement. It will take constitutional skill (at a level thus far not in evidence), and a willingness in Westminster/Whitehall to compromise, if this is to be managed consensually. The process must also be undertaken in a way that fully recognises the special status that Northern Ireland needs to secure, and the special constitutional circumstances that exist.

Promoting Disagreement

Although noted as a mechanism for technical ‘correction’ of EU-originated law, the scope of the Ministerial ‘Henry VIII’ powers, in the absence of a draft bill, are not limited to such a function in practice. There are already significant concerns that the powers will be used with minimal Parliamentary oversight to make sweeping changes. The categories of legislation that will require ‘corrective’ action will in particular be those that reference the exercise of EU-treaty rights that will become redundant post-Brexit. This therefore includes a significant raft of provisions that afford rights to Irish citizens in Northern Ireland, consistent with the GFA context.

In Northern Ireland there is a real prospect of disagreement in the Executive. It is reasonable to foresee Ministers from different parties within a devolved Executive tabling secondary legislation to incorporate EU-originated law that has been ‘corrected’ in such a manner that is an anathema to other parties. For example, in the last mandate the ‘Communities portfolio’, which covers social security and areas of equality law, was held by a DUP Minister who had previously tabled the party’s proposed ‘conscience clause’ private members’ bill.  This legislation sought to legalise GFS discrimination on grounds of sexual orientation when undertaken for reasons of ‘religious conscience’.

Without the prospect of amendment the only option for parties who oppose the manner in which legislation has been ‘corrected’, if it is spotted, is to vote down such legislation, leaving a stalemate situation, whereby EU-originated law is not converted or transposed at all.

The Temptations of ‘Direct Rule’?

One reading of the current situation is that ‘direct rule’ might well suit Westminster/Whitehall. It would avoid some of the complications noted above. In our view, that would hardly be a satisfactory outcome, and it is unlikely, for example, that old-style ‘direct rule’ will be acceptable to either the Irish Government or some of the political parties in Northern Ireland. In the long-term such a scenario involves the GFA political process collapsing on the watch of the UK Government, something successive administrations have sought to avoid.  As we have indicated, Brexit has created a constitutional and political mess for Northern Ireland, and many minds are now puzzling over how to put things back together again. A perfect storm indeed.

Colin Harvey is Professor of Human Rights Law, Queen’s University Belfast and Daniel Holder is Deputy Director of the Committee on the Administration of Justice. This blog is part of the BrexitLawNI project.

(Suggested citation: C. Harvey and D. Holder, ‘The Great Repeal Bill and the Good Friday Agreement – Cementing a Stalemate or Constitutional Collision Course?’, U.K. Const. L. Blog (6th Jun 2017) (available at https://ukconstitutionallaw.org/))