Guest post by Katy Hayward, Queens University Belfast. This article is part of an ongoing blog series ‘Brexit Countdown’ by the Political Studies Association of Ireland (PSAI) and Political Studies Association (PSA).

One of the unwitting effects of the Good Friday (Belfast) Agreement of 1998 is that it potentially put enormous power in the hands of pollsters in Northern Ireland. If there are sure signs of a majority in favour of Irish unity, the British government is required to bring forward legislation to hold a referendum on the matter.

A key means of assessing changing views in Northern Ireland is through attitudinal surveys, the leading example being the Northern Ireland Life and Times Survey (NILT). Every year, NILT poses the question: What do you think the long-term policy Northern Ireland should be? In the 20 years since the 1998 Agreement, the firm preference of the majority of respondents has been ‘devolved government within the UK’, whilst the proportion of those wanting Irish unity has remained on average around 21% (albeit with wide fluctuations).

There have always been differences between the preferences of Catholic and Protestant respondents on this question, but not in quite such a clear-cut way as typically depicted. Although it is rare for Protestant respondents to want Irish unity, it is not unusual for Catholic respondents to see Northern Ireland’s long-term future as being within the UK. This common ground of accepting devolution is, however, conditional on the terms established by the 1998 Agreement.

It is vital for the union to ensure that devolution works. This is seen in the fact that the highest preference for Irish unity recorded by NILT was 30% (inc. 56% of Catholic respondents) in 2006, after 4 years of direct rule. The trend in polarisation of Catholic/Protestant opinions was only reversed by the restoration of the devolved power-sharing institutions in 2007.

In the present day, Brexit is not only coinciding with the absence of a NI Assembly and Executive, it also diminishes trust between the parties and propels them in different ideological directions. Furthermore, it disrupts the context and functioning of relations across all three strands of the Agreement. British-Irish cooperation will have to adjust to operating outside the common environment of the EU. The significance of the Irish border will change in symbolic and practical ways. And the very issue of Brexit is a fierce point of contention between unionists and nationalists.

Will it also see a change in preferences for Northern Ireland’s long-term future? In 2017 (the latest survey for which data is available), NILT asked the direct question: ‘does Brexit make you more in favour of a united Ireland?’ A clear majority of respondents (64%) said it made no difference. There was, however, a community divide on this; 30% of Catholics reported Brexit making them more favourable towards Irish unity, compared to 18% of those of No Religion, and 5% of Protestants.

Perhaps more illuminating is a comparison between 2015 and 2017 on the question of long-term policy for Northern Ireland, the later results having been gathered after the triggering of Article 50. From this we see a significant rise in the proportion of Catholics thinking that Irish unification is the best long-term policy for NI (from 32 to 41%). For Protestant respondents, there is a more complicated trend: a growth in support for devolution, ‘don’t knows’ and ‘other’, plus a very slight increase in preference for a united Ireland (1 to 3%). Among those claiming ‘No Religion’ there is an increase both in those wanting Irish unity and those wanting direct rule in the UK, which suggests growing discomfort in the ‘middle ground’.

Fig.1 What do you think the long-term policy for Northern Ireland should be? (NILT, 2015 & 2017)

Overall, these results indicate that there is no one clear direction of travel in public opinion in Northern Ireland, but rather that opinion is going in differing directions according to community background. The obvious way to manage such divergence peacefully and constructively would be to bolster the effectiveness of the 1998 Agreement institutions. The multi-level governance of Northern Ireland (including British-Irish and cross-border as well as power-sharing dimensions) is the key to its stability – far more so than any simple binary choice of UK vs. united Ireland.

History shows that the stability of Northern Ireland cannot be secured by the firmness of commitment to the union among a majority of people or elected representatives in Northern Ireland. As such, even if there is little evidence of a changing majority to compel the Secretary of State to call a border poll soon, the British government must adhere to its stated commitment (in the Withdrawal Agreement Protocol) to protect the 1998 Agreement in all its dimensions. In the first instance, this means prioritising the restoration of devolution in the interests of Northern Ireland and the union as a whole.

Katy Hayward is a Reader in Sociology and Fellow of the Senator George J. Mitchell Institute for Peace, Security and Justice at Queen’s University Belfast. She is a political sociologist with particular expertise on the impact of European integration on the Irish border and peace process. A version of this blog was presented at the PSA event: ‘Brexit: Towards the Break-up of the United Kingdom?’ held on 24 January 2019 at the British Academy.