Abstract
A popular interpretation of Brexit is as a revolt against the political system or ‘establishment’, yet studies of Brexit in political science have mostly not attempted to model the effects of political discontent on Leave voting. This paper uses data from the British Election Study Internet Panel to probe this in the specific context of the referendum. I present descriptive evidence that Brexit voters were indeed more likely to express attitudes indicating discontent (low satisfaction with democracy, low trust in MPs, and low external efficacy). However, while differences in the levels of discontent between Remainers and Leavers are apparent before the campaign, they widen substantially during it, suggesting a degree of polarisation in response to the negative cues sent out by the Leave campaign about the political system. However, I find no evidence that these factors drove voting on the day, with none of the discontent-related attitudes emerging as significant predictors. While Brexit voters did express more political discontent, it does not appear that this was an important factor in their referendum votes. Rather, I present evidence that the propensity towards anti-system sentiment among Brexiteers was substantially, although not wholly, a product of the referendum campaign itself.
Original language | English |
---|---|
Publication status | Published - Nov 2017 |
Event | Brexit and the Social Sciences (UK in a Changing Europe/ESRC event) - University of Manchester Duration: 23 Nov 2017 → … |
Conference
Conference | Brexit and the Social Sciences (UK in a Changing Europe/ESRC event) |
---|---|
Period | 23/11/17 → … |
Fingerprint
Dive into the research topics of 'Did Brexit express or fuel political discontent?'. Together they form a unique fingerprint.Datasets
-
British Election Study Combined Wave 1-13 Internet Panel
Fieldhouse, E. (Creator), Green, J. (Creator), Evans, G. (Creator), Schmitt, H. (Creator), Van Der Eijk, C. (Creator), Mellon, J. (Creator) & Prosser, C. (Creator), University of Manchester, 1 Aug 2017
DOI: 10.15127/1.293723
Dataset