As the UK negotiates the terms of its departure from the EU, every day its citizens receive an onslaught of claims and counterclaims about the many aspects of the Brexit “deal.” Given the complexity of Brexit negotiations and the heated debate surrounding them, how do citizens decide about what issues are important for them and for the country as a whole? What influences their opinions on Brexit and where do their preferences come from? In our study, we focus on the combination of the three key aspects of Brexit negotiations – issue priorities, material and social considerations – as well as the role that parties play in the formation of preferences about “the best Brexit deal for Britain”.
In our research design we follow a growing body of studies that use conjoint experiments to analyse multidimensional policy preferences (Hainmueller et al. 2013; Hobolt and Leeper 2017). This approach allows examining simultaneously the effect of multiple features of policy deals. Our experiment was fielded online in mid-August 2017 via bmg research UK Omnibus. Each of the 1,512 respondents evaluated different Brexit scenarios comprised of issue priorities (access to the single market and the customs union; reduction in net migration to the UK; maintaining UK global influence, and protecting UK sovereignty), material costs (average annual cost burden for UK taxpayers: £200, £100, £50, £0), and social concerns (support for policy packages aimed at protecting income levels of less than £18,000, less than £26,000, and less than £60,000). Importantly, each Brexit scenario was endorsed by one of the four main parties: Conservative, Labour, Liberal Democrats or SNP.
Using complete randomisation, the conjoint task presented two alternative Brexit negotiation scenarios with a different party sponsor and a different set of attributes in each scenario. The dependent variable was the respondents’ rating of each scenario on a 7-point scale where 1 = “strongly disagree”; 7 = “strongly agree.” An example of the experiment can be viewed here. Figure 1 shows the Average Marginal Conditional Effects (AMCEs) and 95% Confidence Intervals (CIs) for a given attribute value of each independent variable. The points without horizontal lines are reference categories.
To assess the impact of material considerations we tested whether the respondents support a Brexit deal that defends their economic interests. Figure 1 shows that scenarios involving lower costs are significantly more likely to receive higher ratings than a scenario with £200 in personal costs. Similarly, social concerns appear to play an important role in preference formation in regards to the Brexit deal. Respondents view least favourably scenarios that protect the upper income class and this concern for low and very low earners is present irrespective of income level of the respondent. Issue priorities play a relatively weaker role in explaining Brexit deal preferences.
Figure 1. Support for Brexit deal attributes: full sample: AMCEs (with 95% CIs)
To measure the impact of party cues on support for a specific Brexit scenario, respondents have been divided into groups according to the party that they would vote for on the day of the experiment (Figures 2a – 2d). In line with expectations, respondents likely to vote for a specific party tend to support the policy scenario endorsed by that party. Interestingly, we find that while Conservative supporters dislike scenarios endorsed by all other parties, Labour, Liberal Democrats and SNP supporters tend to prefer a policy scenario that is not endorsed by a Conservative sponsor. This is an important finding as it hints at polarisation of UK party politics – while Labour supporters have positive evaluation of the policy offer of their own and other parties, Conservative supporters have negative evaluation of policy offers of all other parties. Thus, party cues have a clear impact on the evaluation of various Brexit scenarios for the supporters of all parties but that impact is greater for Conservative identifiers.
Figure 2a. Support for Brexit deal attributes: Conservative supporters
Figure 2b. Support for Brexit deal attributes: Labour supporters
Figure 2c. Support for Brexit deal attributes: Liberal Democrats supporters
Figure 2d. Support for Brexit deal attributes: SNP supporters
Testing the impact of individual level characteristics, such as education, multiculturalism and support for the EU, we find strong heterogeneity across respondents. Education appears to impact issue priorities quite strongly and to a limited extent costs preferences. Multiculturalism renders similar results for the two groups of respondents except for issue priority. However, the most interesting finding concerns the respondents’ attitudes toward the EU (Figure 3a and 3b). In this case, the way the respondents voted in the Brexit referendum marks clear differences in priority concerns. Although both groups would like to pay as little as possible for Brexit and both agree on income protection for the lowest earners, issue priorities are exactly opposite for Leavers and Remainers. While the reduction of net migration is the most important issue for Leavers, it is the least important issue for Remainers and the other issue priorities are the exact opposite for the two groups. This seems to be another indication of polarisation among the British public, which could mean that it will be very challenging for the government to find a policy compromise that is acceptable for most Britons. In other words, as far as “the best Brexit deal for Britain”, it may not be possible to give everyone what they want.
Figure 3a. Support for Brexit deal attributes: Leave voters
Figure 3b. Support for Brexit deal attributes: Remain voters
Carolina Plescia is Assistant Professor in the Department of Government at the University of Vienna. Her research focuses on public opinion, electoral behaviour and political representation. Magdalena Staniek is a Government of Ireland postdoctoral research fellow in the School of Politics and International Relations, University College Dublin and a researcher at the Higher Education Authority of Ireland. Ideas in this piece were originally presented at the EPOP 2017 conference hosted by the School of Politics and International Relations between 8-10th September 2017. Image credit: CC by bankenverband/Flickr.