Northern Ireland: Power-Sharing in Crisis

Cera Murtagh
Dr Cera Murtagh is Assistant Professor of Comparative and Irish Politics at Villanova University, Pennsylvania, formerly part of the ESRC project Exclusion amid Inclusion: Power-Sharing and Non-Dominant Minorities at Queen’s University Belfast. She has published a number of articles relating to civic mobilization in divided societies, particularly under power-sharing conditions.



In spite of numerous suspensions in its initial years, nationalist and unionist parties shared power for an uninterrupted 10 years from 2007-17. At the time of writing, however, Northern Ireland finds itself in a seemingly intractable political crisis, produced by both internal and external factors, and the future of power-sharing hangs in the balance. The impasse underlines the need for broad inclusion in power-sharing arrangements, beyond the core ethno-national parties. It further speaks to the importance of continued constructive engagement from external actors, who were central to the conflict and remain central to its resolution.



The present moment feels an odd time to be writing about power-sharing in Northern Ireland. Despite the region being catapulted to the centre of a geopolitical crisis following the decision of the United Kingdom to exit the European Union in 2016, the Northern Ireland Assembly and Executive have been suspended for a world-record-beating almost three years. As critical events unfold that could determine Northern Ireland’s future peace and political stability – not to mention its constitutional status – Stormont, the seat of the legislature in Belfast, lies dormant.

What does all of this mean for power-sharing? A superficial reading might suggest simple failure. Yet this would be to overlook the fact that power-sharing sustained in Northern Ireland for ten years from 2007-17. Moreover, the current crisis emerged from a combination of internal and external factors, many of which extended well beyond the scope of its governing arrangements.

Yet, the predicament in which Northern Ireland finds itself does bear lessons for the practice of power-sharing, both here in future and in other divided societies. It underscores the need to forge and maintain broad-based and inclusive power-sharing arrangements that go beyond a focus on the main groups. And it demonstrates the centrality of external actors, not just in creating power-sharing agreements but in sustaining it as it moves forward.



Consociational power-sharing is a model for the governance of deeply divided societies premised on the inclusion of all significant groups in government via their elite representatives. First theorized by Arend Lijphart, consociation was conceived as embodying four conditions: grand coalition government comprising the main groups; proportional representation of those groups; mutual veto for each on matters of critical importance to their community; and group autonomy in certain policy areas (1977). More recent theoretical refinements have included a distinction between corporate consociation – where the groups to share power are pre-determined – and liberal consociation – where those groups are self-determined by the people through the electoral process (McGarry & O’Leary, 2007: 675; McCulloch, 2014: 503). Consociation has increasingly become a tool of external intervention in conflicts (Guelke, 2012; Keil, 2013), prescribed for places as diverse as Northern Ireland, Burundi, Lebanon, Iraq, Bosnia and Herzegovina and Kosovo.

Opponents of the model claim that, far from resolving conflict, by institutionalizing group representation consociation increases polarization in society (c.f. Taylor, 2009). Scholars have pointed to its proneness to dysfunction and deadlock (Horowitz, 2014: 18), while others highlight its exclusionary nature, marginalizing ‘others’ – actors and issues beyond the main ethno-national groups (Stojanović, 2018). The European Court of Human Rights has ruled on three occasions against the state of Bosnia and Herzegovina for discriminating in its electoral laws against citizens beyond Bosniaks, Croats and Serbs.

Proponents, most notably John McGarry and Brendan O’Leary, offer a pragmatic rebuttal: while far from perfect, for many conflict-torn places power-sharing is the only game in town for securing peace and stability. What is more, many of its alleged failings are addressed by liberal power-sharing (McGarry & O’Leary, 2009).


Power-Sharing in Northern Ireland

The solutions embedded in the 1998 Good Friday Agreement were complex and multi-layered. The settlement was characterized by ‘parity of esteem’ between the two political traditions in Northern Ireland (nationalism and unionism), broad inclusion of those communities in the political system, and a confederal, intergovernmental approach with key roles for not only the British but the Irish Government. As well as a newly devolved Assembly and Executive to govern Northern Ireland internally, new institutions were established to facilitate north-south and east-west relations. In this way, power-sharing in Northern Ireland has been referred to as “complex” (Wolff, 2009) and “consociation plus” (O’Leary, 1999; 2018).

But how did power-sharing work in practice? In line with the principles of broad inclusion and proportionality, seats in the Executive were allocated through the d’Hondt formula, based on party strengths in the Assembly. The Executive was, in turn, led by a First Minister and Deputy First Minister of equal authority. The 108 seat Assembly (reduced to 90 after the 2016 elections), was elected through Proportional Representation Single Transferrable Vote (STV), allowing for proportional inclusion of nationalist and unionist (and other) parties (Jarrett, 2018). In order to facilitate group protections, Members of the Legislative Assembly (MLAs) were required to designate in the Assembly as ‘nationalist’, ‘unionist’ or ‘other’. Certain ‘key votes’ (such as financial votes and election of the speaker) required cross-community support, in which the votes of ‘others’ did not count on an equal basis to unionists and nationalists. Cross-community votes could also be triggered by a ‘petition of concern’, signed by 30 Members, an effective ‘community veto’. ‘Group autonomy’ was limited, primarily featuring in education with equal funding for schools of each denomination. In this way, Northern Ireland’s consociation aligned predominantly with the liberal model but, with corporate elements that arguably reinforced the principle of ethnic representation (Nagle, 2016).

After a rocky start with a number of suspensions, a renegotiation at the 2006 St Andrews Agreement (SAA) gave way to uninterrupted power-sharing for ten years (2007-2017), with the heretofore more hardline parties on either side, the Democratic Unionist Party (DUP) and nationalist Sinn Féin, replacing the so-called moderates of the UUP and SDLP.

The institutions evolved significantly after 1998, with four subsequent agreements, most notably St Andrews, which, among other reforms, altered the procedures for election of First and Deputy First Ministers. Initially elected through a cross-community vote, they were subsequently nominated by the largest party in the largest and second largest designations respectively (nationalist, unionist or other) (McEvoy, 2015). The 2014 Stormont House Agreement made provision for an official opposition, which came into effect after the 2016 election.


Power-Sharing in Crisis

So, how did we get here? The power-sharing institutions collapsed in January 2017 when Sinn Féin withdrew from the Executive citing a litany of alleged offences by its unionist partner, most prominently relating to the Renewable Heat Incentive (RHI), scandal in which DUP First Minister Arlene Foster was embroiled. A number of unresolved issues were further listed, including the DUP’s refusal to allow for legislation to protect the Irish language and its use (or ‘abuse’) of the petition of concern to block same-sex marriage in 2015. At the time of writing several rounds of talks have failed to reach agreement to restore the institutions.

The current crisis is underpinned by internal and external conditions. Internally, relations between the power-sharing partners had been under strain for some time. While the early years were dominated by issues of peace and security – most notably weapons decommissioning – over the last ten years, the focus has shifted to issues of communal identity and culture such as parading, flags and – most recently – language. After the broad-based, inclusive process of the GFA, the 2006 St Andrew’s Agreement also saw a narrowing of the agenda and actors, with the focus shifting towards securing sufficient consensus between the DUP and Sinn Féin, the parties that emerged largest in the previous election. That narrowing course continued in the subsequent agreements, which were largely designed to solve power-sharing stalemates. As Michelle Rouse observes “… in each successive negotiation since 1998, equality and human rights elements have been eroded with consistency and power issues aggrandized” (Rouse, 2016: 240).

Nevertheless, power-sharing had faced other crises in the decade leading up to 2017 and managed to stutter forward; what had changed this time were the external conditions. Most dramatically, the decision of the UK in 2016 to leave the European Union served to severely destabilize the arrangement, threatening to undermine the principles of the GFA and with it the peace and security it has built (see Hayward, 2018; Hayward & Murphy, 2018; O’Leary, 2018; Murphy, 2018; Garry et al, 2018). The UK Government’s subsequent pursuit of a so-called ‘hard Brexit’ invoked the possibility of the return to border infrastructure on the island, forcing something of a binary choice with respect to the Agreement’s north-south and east-west dimensions (Hayward, 2018). In so doing, Brexit has reignited the question of Northern Ireland’s constitutional future, heightening insecurity and polarizing politics in the process. To complicate matters further, a ‘confidence and supply’ agreement between the governing Conservative Party and the DUP following the 2017 UK General Election, compromised the UK Government’s legitimacy as guarantor of the Agreement and has vexed its attempts to reach a withdrawal deal with the EU that would avoid a hard border on the island.

After these external shocks, the space between the parties has widened. With their incentives reconfigured, devolved power-sharing in Belfast seems obscured by bigger battles. As a UK General Election approaches at the time of writing, the fate of Brexit – and the fall-out for Northern Ireland – are impossible to predict. The prospects of reestablishing power-sharing while the issue is still live appear slim. Indeed, as Brendan O’Leary warns, even beyond that point, the assumption that the DUP and Sinn Féin will work together within whatever outcome finally emerges from the UK’s negotiations with the EU is still to be tested (2018). Meanwhile, UK Secretary of State for Northern Ireland Julian Smith has warned of fresh elections for Stormont if power-sharing is not restored by mid-January.


Power-Sharing Lessons?

Patently, the blame for the current crisis cannot be simply laid at power-sharing’s door. Significant progress has been achieved since the Agreement, not least police reform and decommissioning. And, despite severe strain, power-sharing endured for ten years up to 2017, a period that saw the transfer of justice powers from Westminster to Stormont. It also witnessed moderating trends: the number and vote share of civic parties increased from approximately 7% in 2007 to 13% in 2017. In 2019 Alliance also won its first seat in the European Parliament and in the local election saw its vote share increase almost 5%, from 6.7% to 11.5%. What’s more, under liberal rules for government formation, open to any party with sufficient support, the largest such party, Alliance, secured seats in the Executive from 2011 to 2016. In addition to holding the Employment and Learning ministry from 2011-16, the Alliance was awarded the contested Justice ministry in 2010 outside of the normal d’Hondt rules. Brexit is an unprecedented external shock that, one might argue, the best designed arrangements would struggle to withstand.

Yet there are lessons to be drawn. Analysis of Northern Ireland’s power-sharing institutions during the 2007-17 period by myself and my colleagues as part of an ESRC research project, points to structures that contributed to polarization, dysfunction and marginalization of ‘other’ issues, identities and actors – including women, LGBT+ groups and smaller ethnic minorities (Agarin et al, 2018).

The structure that most obviously discriminates against ‘others’ and strays from the liberal power-sharing model, is designation. The rule disempowers parties that try to reach across the divide, like Alliance and the Greens, limiting their role in the institutions (Murtagh, 2018). Advocates call for its replacement with designation-free ‘weighted majority’ voting on issues requiring cross-community consent.

The ‘petition of concern’, central to the current dispute, is also deeply problematic. The ‘permissive’ veto – not restricted to specific criteria (McCulloch, 2018) – has been used on a number of occasions, by parties on both sides, on matters that do not directly relate to community rights, including by the DUP to block marriage equality. One representative of the LGBT+ sector thus described the power-sharing system as one that is “not created to effect change. It’s created to slow everything down” (Murtagh & McCulloch, 2018).

Research by Bernadette Hayes and John Nagle highlights the ways in which the veto’s use on matters of broader equality and human rights contributes to ethno-national conflation and contestation of these issues (2016). The resulting deadlock can exacerbate polarization, dysfunction and instability. A more insidious effect, observed in our research, is its stymying effect on debate. Its very existence has arguably shut down discussion on contentious issues like abortion, which might otherwise slowly build towards consensus.

Until October 21st of this year, Northern Ireland stood as the only part of the UK or the island of Ireland where abortion and marriage equality remained illegal. After long and hard-fought campaigns, the legislation to legalise same-sex marriage and decriminalize abortion finally came, not from Stormont but Westminster, passed directly by the UK Parliament.

Other issues highlighted in our research include the procedures for electing First and Deputy First Ministers, which – despite being co-equal – can place the focus of elections on which party prevails, thereby bolstering DUP and Sinn Féin. The insulation of the institutions from civil society also emerged, with advocates calling for the re-establishment of the civic forum established as part of the GFA to sit alongside the Assembly (Pierson & Thomson, 2018; Mackay & Murtagh, 2019).

The story of power-sharing and Northern Ireland’s political crisis could perhaps be read in two ways: as the culmination of years of polarization and failure; or as gradual progress, disrupted. The truth may lie somewhere in between. The most obvious lesson the current plight offers is the need for external actors to remain engaged and play a constructive role. Yet, Northern Ireland’s prior power-sharing experience is also instructive for its future and that of other deeply divided places. That experience speaks to the need for arrangements that, while guaranteeing the rights and security of the groups protected, remain as broadly inclusive as possible.


Suggested Citation: Murtagh, C 2019. ‘Northern Ireland: Power-Sharing in Crisis’. 50 Shades of Federalism. Available at:




Agarin, T., McCulloch, A. & Murtagh, C. (2018) Others in Deeply Divided Societies: A Research Agenda, Nationalism and Ethnic Politics, 24(3): 299-310.

Guelke, A. (2012) Politics in Deeply Divided Societies (Cambridge: Polity Press).

Hayes, B. & Nagle, J. (2016) ‘Ethnonationalism and Attitudes towards Gay and Lesbian Rights in Northern Ireland,’ Nations and Nationalism, 22(1): 20-41.

Horowitz, D. (2014) ‘Ethnic Power Sharing: Three Big Problems,’ Journal of Democracy, 25 (2), 5-20.

Jarrett H (2018) Peace and Ethnic Identity in Northern Ireland: Consociational Power-Sharing and Conflict Management. London: Routledge.

Keil, S. (2013) Multinational Federalism in Bosnia and Herzegovina (Surrey: Ashgate).

Lijphart, A. (1977) Democracy in Plural Societies: A Comparative Exploration (New Haven & London: Yale University Press).

Mackay, F. & Murtagh, C. (2019) ‘New Institutions, New Gender Rules? A Feminist Institutionalist Lens on Women and Power-Sharing,’ feminists@law, 9(1).

McCulloch, A. (2014a) ‘Consociational Settlements in Deeply Divided Societies: The Liberal-Corporate Distinction,’ Democratization, 21(3): 501-518.

McCulloch, A. (2018) ‘The Use and Abuse of Veto Rights in Power-Sharing Systems: Northenr Ireland’s Petition of Concern in Comparative Perspective’, Government and Opposition, 53(4): 735-756.

McEvoy, J. (2015) Power-Sharing Executives: Governing in Bosnia, Macedonia and Northern Ireland (Philadelphia: University of Pennsylvania Press).

McGarry, J. & O’Leary, B. (2007) ‘Iraq’s Constitution of 2005: Liberal Consociation as Political Prescription,’ International Journal of Constitutional Law 5, 4: 670–698.

McGarry, J. & O’Leary B. ‘Power shared after the deaths of thousands,’ in Taylor, R. (2009) (Ed) Consociational Theory: McGarry and O’Leary and the Northern Ireland Conflict (Oxon: Routledge).

Murtagh, C. (2019) ‘The Plight of Civic Parties in Divided Societies,’ International Political Science Review,

Murtagh, C. & McCulloch, A. (2018) ‘Beyond the Core:  The Inclusion of ‘Others’ in Maturing Power-Sharing Systems,’ Paper presented at International Political Science Association World Congress, Brisbane.

Nagle, J. (2016) Social Movements in Violently Divided Societies: Constructing Conflict and Peacebuilding (London: Routledge).

O’Leary, B. (2018) ‘The Twilight of the United Kingdom & Tiocfaidh ár lá: Twenty Years after the Good Friday Agreement,’ Ethnopolitics, 17(3): 223-242.

O’Leary, B (1999) “The 1998 British-Irish Agreement: Power-Sharing Plus,” Scottish Affairs 26(1): 14–35.

Pierson, C. & Thomson, J. (2018) ‘Allies or Opponents? PowerSharing, Civil Society, and Gender,’ Nationalism and Ethnic Politics, 24:1, 100-115.

Rouse, M. (2016) “In need of a fresh start: gender equality in post-GFA Northern Ireland”, NILQ 67 (2); 233-240.

Stojanović, N. (2018) Political marginalization of others in consociational regimes. Zeitschrift für Vergleichende Politikwissenschaft 1–24.

Taylor, R. ‘The Injustice of a Consociational Solution to the Northern Ireland Problem’ in Taylor, R. (2009) (Ed) Consociational Theory: McGarry and O’Leary and the Northern Ireland Conflict (Oxon: Routledge).

Wolff, S. (2009) ‘Complex Power-sharing and the Centrality of Territorial Self-governance in Contemporary Conflict Settlements,’ Ethnopolitics, 8:1, 27-45.


Further Reading


O’Leary, B. (2019) A Treatise on Northern Ireland, Volumes I-III (Oxford: Oxford University Press).

McCulloch, A. and McGarry, J. (eds) (2017) Power-Sharing: Empirical and Normative Challenges (London: Routledge).

Hayward, K. (2018) ‘The pivotal position of the Irish border in the UK’s withdrawal from the European Union’, Space and Polity, 22(2): 238-254.

Hayward, K. & Murphy, M. C. (2018) ‘The EU’s Influence on the Peace Process and Agreement in Northern Ireland in Light of Brexit,’ Ethnopolitics, 17(3): 276-291.

Garry, J., McNichol, K., O’Leary, B. & Pow, J. (2018) Northern Ireland and the UK’s Exit from the EU: What do People Think?, Economic and Social Research Council

Murphy, M. C. (2018) Europe and Northern Ireland’s Future: Negotiating Brexit’s Unique Case (Agenda Publishing).

Tonge, J. (2017) ‘Supplying Confidence or Trouble: The Deal between the Democratic Unionist Party and the Conservative Party,’ Political Quarterly, 88(3): 412-416.

Leave a Reply